To estimate the duration of moderate-to-severe menopausal hot flushes and identify potential risk factors for hot flush duration.
The Penn Ovarian Aging Study cohort was monitored for 13 years. Hot flushes were evaluated at 9-month to 12-month intervals through in-person interviews. The primary outcome was the duration of moderate-to-severe hot flushes estimated by survival analysis (n=259). Potential risk factors included menopausal stage, age, race, reproductive hormone levels, body mass index (BMI), and current smoking. A secondary analysis included women who reported any hot flushes (n=349).
The median duration of moderate-to-severe hot flushes was 10.2 years and was strongly associated with menopausal stage at onset. Hot flushes that started near entry into the menopause transition had a median duration greater than 11.57 years; onset in the early transition stage had a median duration of 7.35 years (95% confidence interval [CI] 4.94–8.89; P<.001); and onset in the late transition to postmenopausal stages had a median duration of 3.84 years (95% CI 1.77–5.52; P<.001). The most common ages at onset of moderate-to-severe hot flushes were 45–49 years (median duration, 8.1 years; 95% CI 5.12–9.28). African American women had a longer duration of hot flushes than white women in adjusted analysis.
The median duration of hot flushes considerably exceeded the timeframe that is generally accepted in clinical practice. The identified risk factors, particularly menopausal stage, race, and BMI, are important to consider in individualizing treatment and evaluating the risk-to-benefit ratio of hormones and other therapies.
The median duration of moderate-to-severe menopausal hot flushes exceeds the generally accepted timeframe and is significantly associated with menopausal stage at onset.
From the Department of Obstetrics/Gynecology, the Center for Clinical Epidemiology and Biostatistics, and the Center for Research in Reproduction and Women's Health, School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania; and the Division of Biostatistics, Department of Medicine, Indiana University, Indianapolis, Indiana.
Supported by grants from the National Institutes of Health: RO1-AG-12745 (E.W.F., Principal Investigator) and RR024134 (Clinical and Translational Research Center).
Corresponding author: Ellen W. Freeman, PhD, Department of Obstetrics/Gynecology, 3701 Market Street, Suite 820 (Mudd), Philadelphia, PA 19104; e-mail: email@example.com.
Financial Disclosure Ellen W. Freeman has received research support from Forest Laboratories, Inc, Wyeth, Pfizer, and Xanodyne Pharmaceuticals. She has also received honoraria for consulting and presentations from Wyeth, Forest Laboratories, Inc, Pherin Pharmaceuticals, and Bayer Health Care. The other authors did not disclose any potential conflicts of interest.