Factors associated with developing vaginal dryness symptoms in women transitioning through menopause: a longitudinal study

Abstract

Objective: To evaluate factors associated with incident self-reported vaginal dryness and the consequences of this symptom across the menopausal transition in a multiracial/ethnic cohort of community-dwelling women.

Methods: We analyzed questionnaire and biomarker data from baseline and 13 approximately annual visits over 17 years (1996-2013) from 2,435 participants in the Study of Women's Health Across the Nation, a prospective cohort study. We used discrete-time Cox proportional-hazards regression to identify predictors of incident vaginal dryness and to evaluate vaginal dryness as a predictor of pain during intercourse and changes in sexual intercourse frequency.

Results: The prevalence of vaginal dryness increased from 19.4% among all women at baseline (ages 42-53 years) to 34.0% at the 13th visit (ages 57-69 years). Advancing menopausal stage, surgical menopause, anxiety, and being married were positively associated with developing vaginal dryness, regardless of partnered sexual activity. For women not using hormone therapy, higher concurrent levels of endogenous estradiol were inversely associated (multivariable-adjusted hazard ratio: 0.94 per 0.5 standard deviation increase, 95% confidence interval: 0.91-0.98). Concurrent testosterone levels, concurrent dehydroepiandrosterone sulfate levels, and longitudinal change in any reproductive hormone were not associated with developing vaginal dryness. Both vaginal dryness and lubricant use were associated with subsequent reporting of pain during intercourse, but not with a decline in intercourse frequency.

Conclusion: In these longitudinal analyses, our data support many clinical observations about the relationship between vaginal dryness, menopause, and pain during intercourse, and suggest that reporting of vaginal dryness is not related to androgen level or sexual intercourse frequency.

Conflict of interest statement

The authors have no conflict of interest to disclose.

Figures

Figure 1. LOESS plot of prevalence of any vaginal dryness in the prior 2 weeks in an 18-year period bracketing the final menstrual period

Figure represents 1593 women (19,119 observations) with the final menstrual period observed prior to initiation of hormone therapy, hysterectomy, or bilateral oophorectomy.

Figure 2

Figure 2a. Multivariable-adjusted hazard ratios for reporting incident pain during intercourse. Multivariable-adjusted hazard ratios for reporting incident pain during intercourse in relation to reporting of vaginal dryness alone, lubricant use alone, vaginal dryness and lubricant use together, and intercourse frequency in SWAN (1996–2013) among 1,474 women who reported having partnered sexual activity in 4,764 visits; excluding women who reported pain during intercourse at baseline and women with missing baseline partner status (and thus missing baseline pain). Models included variable of interest one at a time (for example, vaginal dryness at concurrent visit) and adjusted for age, race and ethnicity, site, menopausal status and hormone use, BMI, anxiety (score ≥ 4), and symptom sensitivity score. Bracket size is proportional to sample size. Figure 2b. Multivariable-adjusted odds ratio for reporting decline in intercourse frequency. Multivariable-adjusted odds ratio for reporting decline in intercourse frequency in relation to vaginal dryness, lubricant use, and pain during intercourse in SWAN (1996–2013) among 2364 women who reported having sexual partners in 13,047 visits. Models included variables of interest one at a time (for example, vaginal dryness at concurrent visit) and adjusted for age at current visit, site, race and ethnicity, years between visits, menopausal status and hormone use, BMI, CES-D (score ≥ 16), smoking, and marital status. Bracket size is proportional to sample size.

Figure 2

Figure 2a. Multivariable-adjusted hazard ratios for reporting incident pain during intercourse. Multivariable-adjusted hazard ratios for reporting incident pain during intercourse in relation to reporting of vaginal dryness alone, lubricant use alone, vaginal dryness and lubricant use together, and intercourse frequency in SWAN (1996–2013) among 1,474 women who reported having partnered sexual activity in 4,764 visits; excluding women who reported pain during intercourse at baseline and women with missing baseline partner status (and thus missing baseline pain). Models included variable of interest one at a time (for example, vaginal dryness at concurrent visit) and adjusted for age, race and ethnicity, site, menopausal status and hormone use, BMI, anxiety (score ≥ 4), and symptom sensitivity score. Bracket size is proportional to sample size. Figure 2b. Multivariable-adjusted odds ratio for reporting decline in intercourse frequency. Multivariable-adjusted odds ratio for reporting decline in intercourse frequency in relation to vaginal dryness, lubricant use, and pain during intercourse in SWAN (1996–2013) among 2364 women who reported having sexual partners in 13,047 visits. Models included variables of interest one at a time (for example, vaginal dryness at concurrent visit) and adjusted for age at current visit, site, race and ethnicity, years between visits, menopausal status and hormone use, BMI, CES-D (score ≥ 16), smoking, and marital status. Bracket size is proportional to sample size.