Associations with menopause and menopausal transition in a nationally representative US sample

Abstract

Objectives: The purpose of this study was to assess sociodemographic and behavioral factors in relation to menopausal status in a representative sample of the United States population. Methods: Data were taken from the 1999 National Health Interview Survey (NHIS), limited to women aged 40–54 years who had not undergone surgical menopause. Menopausal transition was defined as absence of menstrual cycles for at least 3 but no more than 11 months or cycles that had become irregular in the past 12 months. Postmenopause was defined as absence of a menstrual cycle for 12 or more months. We used age-adjusted three-level logistic regression to examine the association between menopausal status and smoking, race/ethnicity, education, body mass index, exercise, and alcohol use. Results: Twenty percent of women in this sample had experienced natural menopause, 18% were in the menopausal transition and 61% were premenopausal. Using premenopause as the reference group, current cigarette smoking was strongly associated with being postmenopausal (odds ratio (OR) 2.2, 95% confidence interval (CI) 1.7, 3.0) and weakly associated with being in the menopausal transition (OR 1.4, 95% CI 1.1, 1.8). Education level was associated with being postmenopausal (OR 0.4, 95% CI 0.3, 0.6 comparing women with a college degree to women who had not completed high school), and alcohol use was weakly associated with being postmenopausal, with no evidence of a dose–response. Conclusions: The associations with smoking were stronger for postmenopause than for the transition phase, suggesting that the effect of smoking may be to shorten the transition period. Education level may be a marker for other exposures that affect ovarian senescence.