Abstract
Objective:
To determine whether reproductive hormones are related to sexual function during the menopausal transition.
Design:
The Study of Women's Health Across the Nation (SWAN) is a multiethnic cohort study of the menopausal transition located at seven US sites. At baseline, the 3302 community-based participants, aged 42–52, had an intact uterus and at least one ovary and were not using exogenous hormones. Participants self-identified as White, Black, Hispanic, Chinese, or Japanese. At baseline and at each of the 10 follow-up visits, sexual function was assessed by self-administered questionnaires, and blood was drawn to assay serum levels of T, estradiol, FSH, SHBG, and dehydroepiandrosterone sulfate.
Main Outcome Measures:
Self-reported frequency of masturbation, sexual desire, sexual arousal, orgasm, and pain during intercourse.
Results:
Masturbation, sexual desire, and arousal were positively associated with T. Masturbation, arousal, and orgasm were negatively associated with FSH. Associations were modest. Estradiol was not related to any measured sexual function domain. Pain with intercourse was not associated with any hormone.
Conclusions:
Reproductive hormones were associated with sexual function in midlife women. T was positively associated, supporting the role of androgens in female sexual function. FSH was negatively associated, supporting the role of menopausal status in female sexual function. The modest associations in this large study suggest that the relationships are subtle and may be of limited clinical significance.
Sexual function is extremely to moderately important to more than 75% of midlife women surveyed (1), and both sexual function and reproductive hormones undergo changes across the menopausal transition. As we reported in the longitudinal study of sexual function in midlife from the Study of Women Across the Nation (SWAN), controlling for a wide range of covariates, stages of the menopausal transition were independently associated with frequency of reported masturbation and with changes in sexual desire and painful intercourse, but not with other components of sexual functioning, including sexual arousal, the frequency of sexual intercourse, physical pleasure with sexual activity, or the importance of sex (2). Although the association with menopausal stage, defined by bleeding criteria, suggests a possible hormonal effect on sexual function, evidence directly associating hormones to sexual function remains inconsistent.
Sex steroids have been postulated to affect sexual function either directly, via neurohormonal effects on desire (3–5), or indirectly by maintaining the anatomic competence that allows arousal to occur (5–7). Because the menopausal transition is a time of change both in endogenous reproductive hormones (8–11) and in sexual function (1, 2, 12–14), a relationship is plausible but is likely to be confounded by the simultaneous effects of normal aging. To date, evidence linking endogenous hormone levels or changes in levels over time to sexual function is mixed, with studies showing small associations with testosterone (T) or estradiol (E2) (13, 15–17) or not supporting a demonstrable relationship (18).
Human female sexual function has been decomposed into the four components of desire, arousal, orgasm, and pain to facilitate research and direct clinical assessment and treatment of female sexual dysfunction (19). Longitudinal studies conducted over the menopausal transition suggest that domains of sexual functioning change differentially over time, with changes in some domains due to aging and changes in others due to menopause (2, 12–14, 20). Few data are available regarding the relationship of endogenous hormones and the domain of masturbation, a marker of sexual function less dependent on a woman's sexual relationship or partner availability (21).
In an extension of the SWAN longitudinal analysis of sexual function with an additional 4 years of follow-up, we evaluated the four domains of desire, arousal, ability to achieve orgasm, and pain with intercourse, as well as the frequency of masturbation, in relation to endogenous levels of reproductive hormones. These included T, E2, FSH, the binding protein SHBG, the adrenal preandrogen dehydroepiandrosterone sulfate (DHEAS), and the calculated estimates of biologically available sex steroids, free T index (FTI) and free E2 index (FEI). To investigate the potentially subtle associations of reproductive hormones with sexual function, we analyzed this large longitudinal dataset to ask the question: Are baseline or concurrent serum levels, or changes in levels, of measured reproductive hormones related to domains of sexual function in midlife women as they transition through the menopause?
Subjects and Methods
Study population
The data are from the participants of SWAN, a multisite, longitudinal cohort study being conducted in community-based groups of women (see Supplemental Data). At baseline, 3302 women who belonged to one of five ethnic/racial groups were recruited: White (n = 1550), Black (n = 935), Japanese (n = 281), Chinese (n = 250), and Hispanic (n = 286). Eligibility criteria for entry into the cohort were: age 42 to 52 years, intact uterus and at least one ovary, no current use of estrogens or other medications known to affect ovarian function, at least one menstrual period in the 3 months before screening, and self-identification with one of the five eligible ethnic groups.
Cohort recruitment and enrollment have been described in detail (22). Briefly, participants were enrolled at seven clinical sites in: Boston, Massachusetts; Chicago, Illinois; the Detroit, Michigan area; Los Angeles, California; Hudson County, New Jersey; Oakland, California; and Pittsburgh, Pennsylvania. Each site enrolled a designated number of White women, whereas the Black women were enrolled in Boston, Chicago, the Detroit area, and Pittsburgh and the Japanese, Chinese, and Hispanic women were enrolled in Los Angeles, Oakland, and Hudson County, NJ, respectively. Institutional Review Board approval was obtained at each study site.
Study Design
All SWAN sites used a single common annual assessment protocol. Data included in this report are from SWAN participants who responded to a self-administered questionnaire on sexual function and who had at least one measure of reproductive hormones from baseline to the 10th annual visit. Sexuality outcome variables were measured at each study visit using a 20-item questionnaire designed to address sexual activity and function using a five- or six-point Likert scale. The questionnaire, which was self-administered and returned to the staff in a sealed envelope, was derived from several sources and has been previously described (2). Variables reflected the domains of sexual desire, masturbation, sexual arousal, ability to climax with a partner, and pain. Women were asked how often they felt desire in the past 6 months to engage in any form of sexual activity either alone or with a partner, how frequently they engaged in masturbation, and if they had engaged in sexual activities with a partner in the last 6 months. Respondents who reported having engaged in sexual activities with a partner were asked about arousal during sexual activity and the ability to climax with a partner. Women who reported having sexual intercourse were also asked about frequency of vaginal or pelvic pain during intercourse and the use of lubricants to facilitate intercourse. In addition to information on partner status, women were asked about their emotional satisfaction with a partner at visits 4–6.
Measured reproductive hormones were the primary independent variables, and included T, E2, FSH, the binding protein SHBG, the adrenal steroid DHEAS, and the calculated estimates of biologically available sex steroids FTI and FEI.
Optimally, women were scheduled for their annual visit, including the sexuality questionnaire, with venipuncture before 10 am on days 2–5 of a spontaneous menstrual cycle. Two attempts were made to obtain the day-2 to -5 sample. If a timed sample could not be obtained, a random fasting sample was taken within a 90-day window of the anniversary of the baseline visit. Blood was refrigerated 1–2 hours after phlebotomy; after centrifugation, the serum was aliquoted, frozen, and batched for shipment to the central laboratory.
Assays
E2 concentrations were measured in duplicate with a modified, off-line ACS-180 (E2-6) immunoassay (Bayer Diagnostics Corporation). Inter- and intra-assay coefficients of variation averaged 10.6 and 6.4%, respectively, over the assay range, and the lower limit of detection was 1 pg/mL. FSH assays were conducted in singlicate using an ACS-180 automated analyzer (Bayer Diagnostics Corporation) and a two-site chemiluminometric immunoassay. Inter- and intra-assay coefficients of variation were 12.0 and 6.0%, respectively, and the lower limit of detection was 1.1 IU/L. The absolute concentrations of FSH are somewhat higher in this assay as compared to values from many clinical laboratories, based on differences in the standards selected. Serum T concentrations were determined by the competitive binding of a dimethyl acridinium ester-labeled T derivative to a rabbit polyclonal antitestosterone antibody premixed with monoclonal antirabbit IgG antibody immobilized on the solid-phase paramagnetic particles. Inter- and intra-assay coefficients of variation were 10.5 and 8.5%, respectively, and the lower limit of detection was 2 ng/dL. The assay for T was standardized analytically by gas chromatography/mass spectroscopy. The de novo two-site chemiluminescent assays for SHBG and DHEAS concentrations involved competitive binding of dimethyl acridinium ester-labeled SHBG or DHEAS to a commercially available rabbit anti-SHBG or anti-DHEAS antibody and a solid phase of goat antirabbit IgG conjugated to paramagnetic particles. Inter- and intra-assay coefficients of variation for SHBG were 9.9 and 6.1%, respectively, and the lower limit of detection was 2 nm. Inter- and intra-assay coefficients of variation for DHEAS were 11.3 and 7.6%, respectively, and the lower limit of detection was 2 μg/dL.
Total T was indexed to SHBG to calculate the FTI: FTI = 100 × T (ng/dL)/28.84 × SHBG (nm). Likewise, total E2 was indexed to SHBG to calculate the FEI: FEI = 100 × E2 (pg/mL)/272.11 × SHBG (nm) (23).
Variables
Menopausal status was based on self-report of bleeding patterns. Bleeding in the 3 months before examination and no decrease in the predictability of menses in the past year was considered premenopausal; bleeding in the past 3 months and a decrease in predictability in the past year was considered early perimenopausal; no menses for 3–11 months was classified as late perimenopausal; and no menses for 12 or more months was categorized as postmenopausal (24). Hot flashes, night sweats, and vaginal dryness were reported as occurring in the 2 weeks before the follow-up visit either not at all, on 1–5 days, or on 6 or more days.
Age was calculated based on the participant's date of birth and the date reported in the questionnaire. Height (centimeters) and weight (kilograms) were measured using a stadiometer and calibrated scales. Body mass index (BMI) was calculated as weight (kilograms)/height (meters)2. Primary race/ethnicity was based on self-definition and reported as Black, White, Chinese, Japanese, and Hispanic. Smoking history was based on seven questions about smoking adapted from the American Thoracic Society standards (25), and women were categorized as never, former, or current smokers. Health status referenced the 12 months before the index visit, and responses were assigned to three categories: worse, same, or better health. Depressive symptoms were assessed by the Center for Epidemiologic Studies Depression Scale (CESD) and categorized as CESD score of ≥16 or not (26).
Analysis
Data were included in this analysis for each study visit where a woman provided information on sexual function, had blood drawn for hormonal analysis, and did not report use of exogenous hormones. At baseline, 3276 (99%) of the 3302 enrollees had questionnaire and hormone data. The New Jersey site had a hiatus in field work from visits 6–10. At follow-up visit 10, 2206 of the original 2870 (76.9%) women from the other six sites completed the self-administrated questionnaire, of whom 1703 (77.2%) contributed both the sexual function questionnaire and hormone data for the analysis. Retention of the eligible cohort was enhanced by recontacting dropouts during follow-up year 9 to re-enter the study, adding 186 participants to the cohort. The highest percentage of dropouts occurred between the baseline evaluation and follow-up visit 1 (13.1%), with dropouts tending to report worse health status, greater BMI, and greater likelihood to be African American or Hispanic.
After examining distributions and crude bivariate analyses, we used generalized estimating equations suitable to estimate population average parameters to build logistic regression models for each of the seven hormones of interest while adjusting for menopausal status, report of hot flashes or night sweats, vaginal dryness, age, time from baseline, availability of a partner, race, health status, BMI, smoking status, difficulty in paying for basics, depressed mood, use of lubricants, and emotional satisfaction with a partner. All visits for each woman were included in the models, except visits when women were using hormone therapy or did not have hormone or sexual activity data. Models included the year of visit to incorporate change with time and, if relevant, an interaction term between hormones and time. Models were developed for each individual hormone, and, based on these findings, additional models were developed for two hormone combinations: 1) FSH, T, E2, SHBG, and DHEAS; and 2) FSH, DHEAS, FTI, and FEI. To explore the possible association of low T levels with sexual dysfunction, models were also built to consider sexual function measures in relation to the upper and lower T quartiles referenced to the middle two quartiles. Nested models were used to test whether hormone associations with sexual function differed by menopausal stage. Finally, stratified models were fit based on the presence or absence of an available partner. Models tested associations of sexual function outcomes with hormone levels at baseline, with change from baseline levels, and with levels concurrent with sexual function responses. Results did not differ when we excluded New Jersey participants. SAS 9.3 and Macro facilities (SAS Institute) were used to perform the statistical analyses.
Results
Population description
Table 1 provides descriptive frequencies at baseline, at the midpoint follow-up visit (visit 5), and the final visit (visit 10). At baseline, all women were premenopausal or early perimenopausal; by follow-up visit 10, 76.9% of respondents were postmenopausal, and 8.5% reported surgical menopause, whereas 128 (6.99%) were using hormone therapy.
Table 1.
Demographic Characteristics of Analyzed Participants at Baseline, Follow-Up Visit 5, and Follow-Up Visit 10: Study of Women's Health Across the Nation
| Baseline | Visit 5a | Visit 10a | |
|---|---|---|---|
| n | 3266 | 1628 | 1703 |
| Age, median (IQR), y | 46 (4) | 51 (4) | 56 (4) |
| BMI, median (IQR), kg/m2 | 26.64 (9.28) | 27.35 (9.78) | 27.68 (9.76) |
| Race/ethnicity | |||
| African American | 920 (28.2) | 443 (27.2) | 511 (30) |
| Caucasian | 1535 (47) | 733 (45) | 824 (48.4) |
| Chinese | 249 (7.6) | 164 (10.1) | 175 (10.3) |
| Hispanic | 285 (8.7) | 97 (6) | N/A |
| Japanese | 277 (8.5) | 191 (11.7) | 193 (11.3) |
| Health status | |||
| Worse | 513 (15.8) | 241 (15) | 247 (14.5) |
| Same | 2028 (62.6) | 522 (32.5) | 584 (34.4) |
| Better | 699 (21.6) | 845 (52.5) | 867 (51.1) |
| Menopausal status | |||
| Pre | 1713 (53.7) | 94 (5.8) | 5 (0.3) |
| Early peri | 1477 (46.3) | 763 (46.9) | 129 (7.6) |
| Late peri | N/A | 227 (14) | 115 (6.8) |
| Post | N/A | 495 (30.4) | 1310 (76.9) |
| Surgical | N/A | 48 (3) | 144 (8.5) |
| Hot flashes | |||
| None | 2382 (73.3) | 847 (52.2) | 837 (49.3) |
| 1–5 d | 617 (19) | 447 (27.5) | 453 (26.7) |
| 6+ d | 252 (7.8) | 330 (20.3) | 407 (24) |
| Night sweats | |||
| None | 2299 (70.7) | 989 (60.9) | 1097 (64.6) |
| 1–5 d | 735 (22.6) | 438 (27) | 378 (22.2) |
| 6+ d | 216 (6.6) | 197 (12.1) | 224 (13.2) |
| Vaginal dryness | |||
| None | 2616 (80.7) | 1122 (69.3) | 1108 (65.3) |
| 1–5 d | 453 (14) | 320 (19.8) | 289 (17) |
| 6+ d | 174 (5.4) | 178 (11) | 301 (17.7) |
| Use of lubricants | |||
| Never | 1888 (73.9) | 708 (63.6) | 417 (43.8) |
| Sometimes/almost never | 435 (17) | 212 (19) | 231 (24.3) |
| Always/almost always | 184 (7.2) | 158 (14.2) | 259 (27.2) |
| No intercourse | 47 (1.8) | 36 (3.2) | 44 (4.6) |
| Overall health | |||
| Fair/poor | 513 (15.8) | 241 (15) | 247 (14.5) |
| Good | 2028 (62.6) | 522 (32.5) | 584 (34.4) |
| Very good | 438 (13.5) | 612 (38.1) | 634 (37.3) |
| Excellent | 261 (8.1) | 233 (14.5) | 233 (13.7) |
| Depression | |||
| Not depressed | 2467 (75.6) | 1311 (80.7) | 1430 (84.4) |
| Depressed | 795 (24.4) | 314 (19.3) | 265 (15.6) |
| Current smoker | |||
| No | 2675 (82.7) | 1423 (87.5) | 1502 (88.4) |
| Yes | 560 (17.3) | 203 (12.5) | 197 (11.6) |
Abbreviations: IQR, interquartile range; N/A, not applicable. Data are expressed as number (percentage) unless otherwise stated.
Women reporting hormone therapy at a follow-up visit were excluded. At baseline, no women were on hormone therapy per protocol. At visit 5, 221 (11.95%) of 1849 women were hormone users; at visit 10, 128 (6.99%) of 1831 women were hormone users.
At baseline, 13.9% of women reported not having a partner (Table 2). By visit 10, 22.8% of women did not have a partner. Over the 10-year period, there was a greater change in the frequency of feeling desire than change in the other domains. At baseline, 58.4% of women reported having desire once or more per week, compared to 35% by visit 10. Other domains of masturbation, arousal, and climax showed much less decline. Slightly less than half of the women (49.4%) reported engaging in masturbation at baseline, compared to 43.8% at visit 10. Approximately two-thirds of women (69.3%) reported feeling aroused during sexual activity with a partner at baseline and 61.1% at visit 10, whereas 88.5% reported that they sometimes or always reached climax at visit 5 and 83.6% at visit 10. In contrast, reporting of pelvic pain at least sometimes increased from 21 to 35% at visit 10.
Table 2.
Self-Reported Sexual Function at Baseline and Follow-Up Visits 5 and 10 in the Study of Women's Health Across the Nation
| Characteristic | Baseline | Visit 5a | Visit 10a |
|---|---|---|---|
| n | 3266 | 1628 | 1703 |
| Availability of a partner | |||
| No partner | 431 (13.9) | 268 (17.3) | 334 (22.8) |
| Partner | 2662 (86.1) | 1279 (82.7) | 1134 (77.2) |
| Frequency of desire to engage in any form of sexual activity, either alone or with a partner(s) | |||
| Less than once per week | 1350 (41.6) | 860 (54.2) | 1066 (64.7) |
| Once or more per week | 1893 (58.4) | 727 (45.8) | 581 (35.3) |
| Frequency of engagement in masturbation (self-stimulation) | |||
| None | 1607 (50.6) | 809 (51.2) | 902 (56.2) |
| Any | 1566 (49.4) | 771 (48.8) | 703 (43.8) |
| Feeling aroused during sexual activity with a partner(s)b | |||
| Sometimes/almost never/never | 785 (30.7) | 401 (36.1) | 370 (38.9) |
| Almost always/always | 1768 (69.3) | 709 (63.9) | 582 (61.1) |
| Reaching climax during sexual activity with a partner(s)b | |||
| Almost never/never | N/A | 128 (11.5) | 155 (16.4) |
| Always/almost always/sometimes | N/A | 982 (88.5) | 793 (83.6) |
| Vaginal or pelvic pain during intercourse with a partner(s)b | |||
| Almost never/never | 1966 (78.8) | 790 (74.2) | 577 (65.1) |
| Always/almost always/sometimes | 530 (21.2) | 275 (25.8) | 309 (34.9) |
Abbreviation: N/A, not applicable. Data are expressed as number (percentage) unless otherwise stated.
Women reporting hormone therapy at a follow-up visit were excluded.
Only women who reported having a partner at a visit answered these questions for that visit.
Endogenous hormones
Median T, FSH, and E2 changed over time with an increase in FSH and decrease in E2 levels (Table 3). The change in T suggested a nonlinear U-shaped pattern. Mean SHBG and DHEAS values did not vary significantly over the 10-year period.
Table 3.
Serum Concentrations of Reproductive Hormones at Baseline and Follow-Up Visits 5 and 10 in the Study of Women's Health Across the Nation
| Characteristics | Baseline | Visit 5a | Visit 10a |
|---|---|---|---|
| n | 3266 | 1628 | 1703 |
| T, ng/dL | 41.5 (26.4) | 35.8 (21.1) | 40.2 (27.1) |
| FSH, mIU/mL | 15.9 (15.6) | 41.45 (68.7) | 108.55 (68.7) |
| E2, pg/mL | 55.2 (55.35) | 25.43 (45.8) | 17.35 (10.95) |
| DHEAS, μg/dL | 114.15 (95) | 113.65 (92.85) | 107.1 (90.7) |
| SHBG, nm | 40.9 (29.4) | 40.5 (32.15) | 39.1 (33.4) |
Data are expressed as median (interquartile range).
Women reporting hormone therapy at a follow-up visit were excluded.
The association between hormone levels and sexual function domains is presented in Table 4. Data are shown only for concurrent hormone levels, because sexual function was most strongly associated with concurrent concentrations, and for the domains of masturbation, desire, and arousal.
Table 4.
Change in Sexual Function Associated With a 1 SD Increase in Concurrent Individual Hormone Levels Adjusted for Demographic and Lifestyle Characteristics, Menopausal Status, and BMI in the Study of Women's Health Across the Nation
| Hormone | Overall |
No Partner |
With a Partner |
||||
|---|---|---|---|---|---|---|---|
| Desire ≥1/wk | Masturbation Ever | Desire ≥1/wk | Masturbation Ever | Desire ≥1/wk | Masturbation Ever | Arousal Almost Always/Always | |
| T, ng/dL | 1.052 (1.015, 1.091)b |
1.073 (1.039, 1.108)d |
1.043 (0.947, 1.149) |
1.112 (1.019, 1.213)a |
1.036 (0.992, 1.081) |
1.074 (1.037, 1.111)d |
1.039 (0.995, 1.085) |
| FSH, mIU/mL | 0.982 (0.938, 1.029) |
0.946 (0.911, 0.983)c |
0.916 (0.811, 1.036) |
0.920 (0.828, 1.022) |
0.991 (0.937, 1.049) |
0.957 (0.916, 1.001) |
0.925 (0.873, 0.981)b |
| E2, pg/mL | 0.994 (0.966, 1.022) |
1.022 (0.999, 1.046) |
0.973 (0.913, 1.036) |
1.044 (0.971, 1.122) |
0.996 (0.960, 1.033) |
1.021 (0.996, 1.047) |
1.014 (0.975, 1.054) |
| DHEAS, μg/dL | 1.058 (1.013, 1.105)a |
1.046 (1.006, 1.087)a |
1.158 (1.046, 1.282)b |
1.076 (0.980, 1.182) |
1.039 (0.987, 1.094) |
1.025 (0.980, 1.072) |
1.051 (0.998, 1.107) |
| SHBG, nm | 0.990 (0.952, 1.029) |
1.028 (0.990, 1.068) |
0.965 (0.874, 1.064) |
1.059 (0.973, 1.153) |
1.001 (0.955, 1.048) |
1.025 (0.982, 1.070) |
0.982 (0.935, 1.031) |
Data are expressed as adjusted odds ratio (95% confidence limit). Odds ratios correspond to 1-SD increments in individual hormone models. In addition to concurrent, baseline and change from baseline for all hormones, tertiles/quartiles of T were also evaluated; data are not reported in this table. Means (SD) in the overall models: T, 40.39 ng/dL (22.33); FSH, 52.25 mIU/mL (50.07); E2, 60.92 pg/mL (79.17); DHEAS, 131.21 μg/dL (81.00); and SHBG, 45.49 nm (26.46).
P < .05;
P < .01;
P < .001;
P < .0001.
Masturbation
Frequency of masturbation was positively associated with T in the individual hormone models, both in the overall cohort and in women with and without a partner. Masturbation frequency was positively associated with the upper quartile of T and negatively associated with the lower quartile of T in the overall cohort and in the subcohorts with and without a partner (data not shown). Frequency of masturbation was negatively associated with FSH and positively associated with DHEAS, but not with E2 or SHBG. Women without a partner reported any masturbation significantly more often (55.44%) than women with a partner (47.23%).
Desire
Frequency of sexual desire was positively associated with T and negatively associated with the lower quartile of T. Desire was positively associated with DHEAS. Frequency of desire was not associated with FSH, E2, or SHBG.
Arousal
Frequency of arousal was negatively associated with the lower quartile of T. Arousal was negatively associated with FSH but was not associated with E2, DHEAS, or SHBG.
Pain with intercourse
The frequency of pain with intercourse, although positively associated with baseline FSH, was no longer significantly associated if the use of lubricants was included in the model. Pain was not associated with any other hormone, including E2 (data not shown).
Climax
The ability to climax with a partner was negatively associated with FSH, but with no other measured hormone (data not shown).
Combined models
Table 5 presents the model that includes all hormones and selected covariates. Significant associations with hormones did not differ, except for masturbation and DHEAS.
Table 5.
Change in Sexual Function Associated With a 1-SD Increase in Concurrent Hormone Levels, Adjusted for All Hormones, Demographic and Lifestyle Characteristics, Menopausal Status, and BMI in the Study of Women's Health Across the Nation
| Parameter | Overall |
No Partner |
With a Partner |
||||
|---|---|---|---|---|---|---|---|
| Desire ≥1/wk | Masturbation Ever | Desire ≥1/wk | Masturbation Ever | Desire ≥1/wk | Masturbation Ever | Arousal Almost Always/Always | |
| T, ng/dL | 1.045 (1.006, 1.085)a |
1.067 (1.033, 1.102)d |
1.022 (0.923, 1.132) |
1.105 (1.009, 1.210)a |
1.030 (0.984, 1.077) |
1.071 (1.034, 1.109)c |
1.028 (0.984, 1.075) |
| FSH, mIU/mL | 0.974 (0.926, 1.025) |
0.951 (0.912, 0.991)a |
0.885 (0.773, 1.012) |
0.928 (0.826, 1.041) |
0.987 (0.928, 1.050) |
0.964 (0.919, 1.011) |
0.922 (0.867, 0.981)a |
| E2, pg/mL | 0.984 (0.953, 1.016) |
1.001 (0.976, 1.027) |
0.947 (0.878, 1.020) |
1.010 (0.939, 1.085) |
0.989 (0.949, 1.030) |
1.002 (0.976, 1.030) |
0.996 (0.955, 1.039) |
| DHEAS, μg/dL | 1.044 (0.998, 1.093) |
1.031 (0.991, 1.073) |
1.147 (1.033, 1.273)a |
1.052 (0.956, 1.158) |
1.031 (0.976, 1.088) |
1.009 (0.964, 1.056) |
1.044 (0.990, 1.101) |
| SHBG, nm | 0.998 (0.959, 1.038) |
1.026 (0.987, 1.067) |
0.985 (0.891, 1.088) |
1.061 (0.972, 1.157) |
1.006 (0.960, 1.056) |
1.023 (0.979, 1.068) |
0.982 (0.934, 1.033) |
| Menopausal status (ref: pre) | |||||||
| Early peri | 0.975 (0.885, 1.075) |
1.069 (0.983, 1.162) |
1.119 (0.866, 1.445) |
0.903 (0.724, 1.125) |
0.977 (0.869, 1.099) |
1.083 (0.985, 1.191) |
0.924 (0.818, 1.044) |
| Late peri | 0.829 (0.712, 0.967)a |
1.051 (0.922, 1.198) |
1.042 (0.693, 1.567) |
0.996 (0.706, 1.406) |
0.852 (0.706, 1.029) |
1.048 (0.902, 1.218) |
0.960 (0.788, 1.169) |
| Post | 0.655 (0.560, 0.766)d |
0.948 (0.830, 1.083) |
0.732 (0.491, 1.092) |
0.843 (0.592, 1.200) |
0.680 (0.560, 0.825)d |
0.937 (0.802, 1.094) |
0.641 (0.525, 0.783)d |
| Surgical | 0.849 (0.677, 1.066) |
0.893 (0.699, 1.142) |
1.173 (0.653, 2.106) |
0.453 (0.251, 0.818)b |
0.898 (0.688, 1.174) |
1.146 (0.873, 1.505) |
0.688 (0.495, 0.957)a |
| Hot flashes (ref: none) | |||||||
| 1∼5 d | 0.987 (0.914, 1.067) |
0.948 (0.889, 1.011) |
1.127 (0.911, 1.394) |
0.951 (0.797, 1.134) |
0.942 (0.859, 1.034) |
0.927 (0.859, 1.000) |
1.019 (0.921, 1.129) |
| 6 + d | 0.957 (0.857, 1.070) |
0.916 (0.831, 1.010) |
0.967 (0.680, 1.375) |
0.996 (0.764, 1.299) |
0.913 (0.798, 1.046) |
0.900 (0.807, 1.003) |
0.959 (0.831, 1.108) |
| Night sweats (ref: none) | |||||||
| 1∼5 d | 0.974 (0.901, 1.053) |
1.014 (0.950, 1.082) |
0.826 (0.669, 1.019) |
1.042 (0.901, 1.204) |
1.035 (0.943, 1.137) |
1.026 (0.950, 1.107) |
1.068 (0.966, 1.179) |
| 6 + d | 0.934 (0.820, 1.064) |
0.990 (0.892, 1.098) |
0.991 (0.678, 1.449) |
1.172 (0.853, 1.612) |
0.944 (0.810, 1.099) |
0.955 (0.848, 1.075) |
1.030 (0.879, 1.206) |
| Vaginal dryness (ref: none) | |||||||
| 1∼5 d | 1.040 (0.955, 1.133) |
1.108 (1.032, 1.189)b |
1.043 (0.780, 1.395) |
1.120 (0.893, 1.406) |
0.963 (0.872, 1.062) |
1.119 (1.038, 1.206)b |
0.840 (0.758, 0.931)b |
| 6 + d | 0.904 (0.798, 1.025) |
0.990 (0.883, 1.111) |
1.112 (0.717, 1.725) |
0.988 (0.707, 1.381) |
0.844 (0.734, 0.972)a |
0.990 (0.876, 1.119) |
0.657 (0.569, 0.770)d |
| Baseline age | 0.958 (0.938, 0.979)d |
0.949 (0.927, 0.972)d |
0.977 (0.929, 1.027) |
0.911 (0.866, 0.957)c |
0.965 (0.943, 0.989)b |
0.955 (0.930, 0.980)c |
0.959 (0.934, 0.985)b |
| Duration of follow-up | 0.950 (0.934, 0.966)d |
1.001 (0.987, 1.016) |
0.979 (0.939, 1.0200) |
0.981 (0.944, 1.018) |
0.946 (0.927, 0.965)d |
1.014 (0.997, 1.030) |
1.004 (0.983, 1.025) |
| Emotional satisfaction with a partner (ref: not at all) | |||||||
| Extreme | N/A | N/A | N/A | N/A | 4.719 (3.647, 6.107)d |
0.748 (0.590, 0.949)a |
12.085 (9.136, 15.986)d |
| Very | N/A | N/A | N/A | N/A | 3.264 (2.564, 4.154)d |
0.800 (0.637, 1.006) |
7.581 (5.895, 9.749)d |
| Moderate | N/A | N/A | N/A | N/A | 2.203 (1.740, 2.789)d |
0.894 (0.712, 1.122) |
3.128 (2.454, 3.988)d |
| Slight | 1.410 (1.111, 1.788)b |
1.068 (0.845, 1.349) |
1.635 (1.276, 2.093)d |
||||
Abbreviation: N/A, not applicable. Data are expressed as odds ratio (95% confidence limit). Odds ratios correspond to 1-SD increment in each current hormone. The hormones were centered at the corresponding grand means. Means (SD) in the overall models: T, 40.39 ng/dL (22.33); FSH, 52.25 mIU/mL (50.07); E2, 60.92 pg/mL (79.17); DHEAS, 131.21 μg/dL (81.00); and SHBG, 45.49 nm (26.46).
P < .05;
P < .01;
P < .001;
P < .0001.
Discussion
These data, from a large and ethnically diverse longitudinal study of midlife women traversing the menopausal transition, demonstrate significant associations between sexual function and the reproductive hormones T, FSH, and DHEAS. The modest associations seen in such a large dataset suggest that the relationships are subtle but generally consistent with proposed models. The most robust and diverse associations observed were with masturbation, a sexual function domain not dependent on partner status and potentially a more sensitive measure of association with hormones. Specifically, masturbation frequency increased with increasing levels of T and DHEAS, and was negatively associated with FSH. Somewhat surprisingly, E2 was not associated with any sexual function domain, including pain with intercourse.
Studies to date of sexual function and endogenous hormones have been mixed. The cross-sectional study by Davis et al (18) of concurrent androgen levels and sexual function in women ages 18–75 did not document a measurable relationship between any hormone concentration and any domain of sexual activity. Longitudinal evidence linking measured reproductive hormones to the female sexual function domains (19) has been varied despite documentation of change in both endogenous hormones (8–11) and sexual function (1, 2, 12–14) across the menopausal transition. Dennerstein et al (15) showed that sexual responsivity and libido were related to a decline in E2, but not androgens, and that dyspareunia was related to E2 levels, whereas Gracia et al (16) noted a negative association of sexual function only with DHEAS and not T, E2, FSH, or LH. Woods et al (17) reported a negative association of urinary FSH and a positive association of urinary estrone glucuronide and T with sexual desire. Given the modest associations between hormones and sexual function noted in SWAN, prior reports may have had insufficient power to detect such differences. Reported associations with estrogen not seen in our study may have been a result of the specific estrogen measured, the protocol for collecting or analyzing it, or differences in the sexual function instruments utilized in the study.
The relationship between masturbation frequency and endogenous reproductive hormones in women, both premenopausal and postmenopausal, has been rarely reported, with Hulter and Lundberg (21) noting a positive cross-sectional correlation with T levels in women with hypothalamic-pituitary-ovarian dysfunction. We found that masturbation, the sexual function arguably least influenced by partner characteristics and availability, was associated with T, FSH, and DHEAS. The consistent associations of masturbation with these hormones in the expected directions suggest that the long-suspected effects of reproductive hormones on sexual function are masked by the influence of relationship status and other psychosocial variables associated with partnered sexual activity. Masturbation was associated with all measures of T, including positive and negative associations in women with the highest and lowest T levels, respectively, supporting a role for T in nonpartnered sex as well as partnered sex. Moreover, these associations largely persisted regardless of partner availability, suggesting that relationship status may play a stronger role in partnered sex than nonpartnered sex, even within a relationship.
Desire, the other measured sexual function domain not entirely dependent on partner availability, was associated only with T and DHEAS, not FSH, E2, or SHBG, supporting the concept that androgens play a role in sexual desire in women. The negative association of desire with T in the lowest quartile, similar to that seen with masturbation, suggests the possibility of a dose effect of endogenous T on sexual function. However, the decline in desire across the menopausal transition is not paralleled by a similar decline in T, leaving the relative importance of T on desire an open question, with other covariates such a vaginal dryness and emotional satisfaction with a partner having more robust associations (2) and likely to play a larger role in a complex functional domain. The U-shaped pattern of T over time is of unclear significance because most longitudinal studies have reported a decline in T until the final menstrual period, then stability. Our data are consistent with those reported by Overlie et al (27), although that study only reported hormone levels through 2 years after final menstrual period.
The partnered sexual function domains of arousal, pain with intercourse, and climax with a partner demonstrated fewer associations with reproductive hormone levels, which is not surprising considering the complex characteristics of a sexual relationship and the effects of a partner's health and sexual functioning. Frequency of arousal was negatively associated with T only in women in the lowest quartile of T, suggesting that low T concentrations may play a role in dysfunctional sexual responsiveness in women. Both arousal and climax were negatively associated with FSH, supporting a potentially independent menopause effect on sexual function independent of age. Somewhat surprisingly, E2 was not associated with pain with intercourse, despite the documented effects of estrogen on vulvovaginal integrity (28). This must be interpreted with caution, however, because the early follicular levels measured in the still-menstruating women do not reflect the higher levels achieved later in a cycle and may not represent the total estrogen exposure and possible estrogen effects on sexual function.
It is equally surprising that E2 was not associated with any of the other major sexual function domains in either single or combined models despite a menopause stage effect in all domains, whether or not the concurrent menopausal status variable was included in the model. This observed menopause effect builds on the report by Avis et al (2) documenting in this same SWAN cohort the independent effects of menopause stage and age on the major domains of sexual function. Moreover, FSH, arguably a better hormonal marker of progression through the menopausal transition than E2 (29), was associated in the combined model with only masturbation, arousal, and orgasm, not desire or pain, suggesting that the effects of the menopausal transition on sexual function may not be mediated directly through sex steroid pathways.
There are several limitations associated with this report. Analytically, the models used incorporated a large number of variables measured over time with many dimensions of physiological and social constructs that do not facilitate simple statistical modeling. Thus, the multiple models tested in a large sample provide an opportunity for identifying statistical significance, whereas the biological effects are minimal. When SWAN was initiated, sexual function instruments validated for epidemiological studies, especially in multiethnic groups, were unavailable, necessitating the adaptation of previously developed questionnaires used in cross-sectional studies. As a result, the SWAN instrument includes no measures of intensity, and measures are limited to recall of only 2 weeks before the annual visit, precluding comparison to more recently developed instruments (1). Women who were obese, who smoked, or who had depressed mood, and women in poor health were more likely to be lost to follow-up. Although we adjusted for these characteristics in the multiple regression analyses, bias may still be present. The menopausal status of women with polycystic ovary syndrome may be misclassified. Despite these concerns, the size of the dataset, the longitudinal design, and the comprehensive analyses provide the best evidence to date of a relationship between reproductive hormones and sexual function.
In summary, we have demonstrated significant associations, albeit modest, between endogenous reproductive hormones and sexual function in midlife women traversing the menopausal transition. Masturbation (the sexual function domain least dependent on partner status), desire, and arousal were positively associated with T, supporting the role of androgens in female sexual function. Masturbation, arousal, and the ability to climax with a partner were negatively associated with FSH, supporting an independent effect of menopausal status on female sexual function. E2 was not related to any measured sexual function domain. However, the modest associations seen in this large and ethnically diverse longitudinal study suggest that the relationships are subtle and may be of limited clinical significance in domains of sexual function also strongly related to partner and psychosocial parameters.
Acknowledgments
We thank the study staff at each site and all the women who participated in the Study of Women's Health Across the Nation (SWAN).
This publication was supported in part by the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health through UCSF-CTSI Grant UL1 RR024131.
The Study of Women's Health Across the Nation (SWAN) has grant support from the National Institutes of Health (NIH), Department of Health and Human Services, through the National Institute on Aging (NIA), the National Institute of Nursing Research (NINR) and the NIH Office of Research on Women's Health (ORWH) (Grants U01NR004061, U01AG012505, U01AG012535, U01AG012531, U01AG012539, U01AG012546, U01AG012553, U01AG012554, and U01AG012495). The content of this article {OR: paper, manuscript, abstract, other – please specify} is solely the responsibility of the authors and does not necessarily represent the official views of the NIA, NINR, ORWH, or the NIH.
Clinical Centers: University of Michigan, Ann Arbor—Siobán Harlow, Principal Investigator (PI), 2011 to present; MaryFran Sowers, PI, 1994–2011; Massachusetts General Hospital, Boston, MA—Joel Finkelstein, PI, 1999 to present; Robert Neer, PI, 1994–1999; Rush University, Rush University Medical Center, Chicago, IL—Howard Kravitz, PI, 2009 to present; Lynda Powell, PI, 1994–2009; University of California, Davis/Kaiser—Ellen Gold, PI; University of California, Los Angeles—Gail Greendale, PI; Albert Einstein College of Medicine, Bronx, NY—Carol Derby, PI, 2011 to present; Rachel Wildman, PI, 2010–2011; Nanette Santoro, PI, 2004–2010; University of Medicine and Dentistry-New Jersey Medical School, Newark—Gerson Weiss, PI, 1994–2004; and the University of Pittsburgh, Pittsburgh, PA—Karen Matthews, PI.
NIH Program Office: National Institute on Aging, Bethesda, MD—Winifred Rossi, 2012 to present; Sherry Sherman, 1994–2012; Marcia Ory, 1994–2001; National Institute of Nursing Research, Bethesda, MD—Program Officers.
Central Laboratory: University of Michigan, Ann Arbor—Daniel McConnell (Central Ligand Assay Satellite Services).
Coordinating Center: University of Pittsburgh, Pittsburgh, PA—Maria Mori Brooks, PI, 2012 to present; Kim Sutton-Tyrrell, PI, 2001–2012; New England Research Institutes, Watertown, MA—Sonja McKinlay, PI, 1995–2001.
Steering Committee: Susan Johnson, Current Chair; Chris Gallagher, Former Chair.
Disclosure Summary: The authors have nothing to disclose.
Footnotes
- BMI
- body mass index
- DHEAS
- dehydroepiandrosterone sulfate
- E2
- estradiol
- FEI
- free E2 index
- FTI
- free T index.
References
- 1. Cain VS, Johannes CB, Avis NE, et al. Sexual functioning and practices in a multi-ethnic study of midlife women: baseline results from SWAN. J Sex Res. 2003;40(3):266–276. [DOI] [PubMed] [Google Scholar]
- 2. Avis NE, Brockwell S, Randolph JF, et al. Longitudinal changes in sexual functioning as women transition through menopause: results from the Study of Women's Health Across the Nation. Menopause. 2009;16(3):442–452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Walling M, Andersen BL, Johnson SR. Hormonal replacement therapy for postmenopausal women: a review of sexual outcomes and related gynecologic effects. Arch Sex Behav. 1990;19:119–137. [DOI] [PubMed] [Google Scholar]
- 4. Kane FJ, Jr, Lipton MA, Ewing JA. Hormonal influences in female sexual response. Arch Gen Psychiatry. 1969;20:202–209. [DOI] [PubMed] [Google Scholar]
- 5. Mooradian AD, Greiff V. Sexuality in older women. Arch Intern Med. 1990;150:1033–1038. [PubMed] [Google Scholar]
- 6. Goldstein MK, Teng NN. Gynecologic factors in sexual dysfunction of the older woman. Clin Geriatr Med. 1991;7:41–61. [PubMed] [Google Scholar]
- 7. Sarrel PM. Sexuality and menopause. Obstet Gynecol. 1990;75:26S–30S; discussion 31S–35S. [PubMed] [Google Scholar]
- 8. Burger HG, Dudley EC, Hopper JL, et al. Prospectively measured levels of serum follicle-stimulating hormone, estradiol, and the dimeric inhibins during the menopausal transition in a population-based cohort of women. J Clin Endocrinol Metab. 1999;84:4025–4030. [DOI] [PubMed] [Google Scholar]
- 9. Burger HG, Dudley EC, Cui J, Dennerstein L, Hopper JL. A prospective longitudinal study of serum testosterone, dehydroepiandrosterone sulfate, and sex hormone-binding globulin levels through the menopause transition. J Clin Endocrinol Metab. 2000;85:2832–2838. [DOI] [PubMed] [Google Scholar]
- 10. Randolph JF, Jr, Sowers M, Bondarenko IV, Harlow SD, Luborsky JL, Little RJ. Change in estradiol and follicle-stimulating hormone across the early menopausal transition: effects of ethnicity and age. J Clin Endocrinol Metab. 2004;89:1555–1561. [DOI] [PubMed] [Google Scholar]
- 11. Randolph JF, Jr, Zheng H, Sowers MR, et al. Change in follicle-stimulating hormone and estradiol across the menopausal transition: effect of age at the final menstrual period. J Clin Endocrinol Metab. 2011;96:746–754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Dennerstein L, Dudley E, Burger H. Are changes in sexual functioning during midlife due to aging or menopause? Fertil Steril. 2001;76:456–460. [DOI] [PubMed] [Google Scholar]
- 13. Dennerstein L, Randolph J, Taffe J, Dudley E, Burger H. Hormones, mood, sexuality, and the menopausal transition. Fertil Steril. 2002;77(suppl 4):S42—S48. [DOI] [PubMed] [Google Scholar]
- 14. Avis NE, Zhao X, Johannes CB, Ory M, Brockwell S, Greendale GA. Correlates of sexual function among multi-ethnic middle-aged women: results from the Study of Women's Health Across the Nation (SWAN). Menopause. 2005;12(4):385–398. [DOI] [PubMed] [Google Scholar]
- 15. Dennerstein L, Lehert P, Burger H. The relative effects of hormones and relationship factors on sexual function of women through the natural menopausal transition. Fertil Steril. 2005;84(1):174–180. [DOI] [PubMed] [Google Scholar]
- 16. Gracia CR, Freeman EW, Sammel MD, Lin H, Mogul M. Hormones and sexuality during transition to menopause. Obstet Gynecol. 2007;109(4):831–840. [DOI] [PubMed] [Google Scholar]
- 17. Woods NF, Mitchell ES, Smith-Di Julio K. Sexual desire during the menopausal transition and early postmenopause: observations from the Seattle Midlife Women's Health Study. J Womens Health (Larchmt). 2010;19(2):209–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Davis SR, Davison SL, Donath S, Bell RJ. Circulating androgen levels and self-reported sexual function in women. JAMA. 2005;294(1):91–96. [DOI] [PubMed] [Google Scholar]
- 19. Basson R, Berman J, Burnett A, et al. Report of the International Consensus Development Conference on female sexual dysfunction: definitions and classifications. J Urol. 2000;163:888–893. [PubMed] [Google Scholar]
- 20. Gracia CR, Sammel MD, Freeman EW, Liu L, Hollander L, Nelson DB. Predictors of decreased libido in women during the late reproductive years. Menopause. 2004;11(2):144–150. [DOI] [PubMed] [Google Scholar]
- 21. Hulter B, Lundberg PO. Sexual function in women with hypothalamo-pituitary disorders. Arch Sex Behav. 1994;23(2):171–183. [DOI] [PubMed] [Google Scholar]
- 22. Sowers MF, Crawford S, Morgenstein D. Design and sampling methods of SWAN: a multi-center, multi-ethnic, community-based cohort study of women and the menopausal transition. In: Wren J., Lobo RA, Kelsey J, Marcus R, eds. Menopause: Biology and Pathobiology. Waltham, Massachusetts: Academic Press; 1999. [Google Scholar]
- 23. van den Beld AW, de Jong FH, Grobbee DE, Pols HA, Lamberts SW. Measures of bioavailable serum testosterone and estradiol and their relationships with muscle strength, bone density, and body composition in elderly men. J Clin Endocrinol Metab. 2000;85:3276–3282. [DOI] [PubMed] [Google Scholar]
- 24. Brambilla DJ, McKinlay SM, Johannes CB. Defining the perimenopause for application in epidemiologic investigations. Am J Epidemiol. 1994;140(12):1091–1095. [DOI] [PubMed] [Google Scholar]
- 25. Ferris B. Epidemiology standardization project (American Thoracic Society). Am Rev Respir Dis. 1978;118:1–120. [PubMed] [Google Scholar]
- 26. Radloff L. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401. [Google Scholar]
- 27. Overlie I, Moen MH, Morkrid L, Skjaeraasen JS, Holte A. The endocrine transition around menopause–a five years prospective study with profiles of gonadotropines, estrogens, androgens and SHBG among healthy women. Acta Obstet Gynecol Scand. 1999;78(7):642–647. [PubMed] [Google Scholar]
- 28. Semmens JP, Tsai CC, Semmens EC, Loadholt CB. Effects of estrogen therapy on vaginal physiology during menopause. Obstet Gynecol. 1985;66(1):15–18. [PubMed] [Google Scholar]
- 29. Randolph JF, Jr, Sowers M, Bondarenko I, et al. The relationship of longitudinal change in reproductive hormones and vasomotor symptoms during the menopausal transition. J Clin Endocrinol Metab. 2005;90(11):6106–6112. [DOI] [PubMed] [Google Scholar]