Skip to main content

01-04-2013 | Gynaecology | Article

New model estimates final menstrual period in menopause


Free abstract

medwireNews: US researchers have developed a model for estimating the timing of the final menstrual period (FMP), giving reassurance and the possibility of clinical guidance on preventative interventions for at-risk menopausal women.

"Some researchers have proposed that an intervention begun 1 or 2 years before the FMP would greatly decrease future fracture risk by preventing the very rapid bone loss that occurs in the few years before and few years after the final menses," said lead study author Gail Greendale (David Geffen School of Medicine at the University of California, Los Angeles) in an associated press release.

Currently, physicians monitor bleeding patterns to evaluate menopausal stage, but this has been considered as an imprecise predictor of when the FMP will occur. Over 60% of women who are classified as early perimenopausal become postmenopausal without any additional clinical bleeding.

The findings, published in The Journal of Clinical Endocrinology and Metabolism, may be "the first step to developing web-based calculators and other tools women can use to estimate where they are in the menopause transition and how far away their final period is," according to Greendale.

The researchers developed a formula for estimating the FMP using current estradiol and follicle-stimulating hormone (FSH) levels, in addition to menopause transition stage, race/ethnicity, and cycle stage during blood sample collection.

The model was tested using longitudinal data from 554 women taking part in the Study of Women's Health Across the Nation. Using current and prior values of estradiol and FSH, the team was able to assess whether women were 2 years prior to the FMP, 1 year prior to the FMP, or at the FMP.

Classifying women as having crossed a FMP-related landmark was possible with high sensitivity and specificity. Indeed, when women were classified as having crossed the 2-year period prior to the FMP, sensitivity was 85% and specificity was 77% when predicted probability exceeded 0.3 (studied range 0.3-0.6).

The model was also able to address women caught between two FMP landmarks; using a probability threshold set to below 0.3 was able to classify women as not having crossed a later landmark with a sensitivity of 92% and a specificity of 80%.

The researchers comment: "Because sensitivity and specificity were presented for a range of thresholds, the user can assign the preferred level of certainty that designates the test as positive and then assess performance characteristics at that certainty."

By Ingrid Grasmo, medwireNews Reporter

Related topics