|Home | About | Journals | Submit | Contact Us | Français|
The primary objective of this study was to utilize the Vulvovaginal Symptom Questionnaire (VSQ) to estimate the prevalence and examine the emotional, life style and sexual impact of vulvovaginal symptoms in postmenopausal women.
We administered the VSQ questionnaire, a previous validated instrument to 358 postmenopausal women recruited from primary care physician offices and local senior centers. The first 7 questions of the VSQ comprise the symptom subscale (itching, burning, hurting, irritation, dryness, discharge and odor). Women who answered “Yes” to any of the first 7 symptom questions were considered to have vulvovaginal symptoms.
Two hundred seventy nine women were recruited from primary care offices and 79 women were recruited from senior centers. One hundred eighty-three post-menopausal women (51.1%; 95% CI 45.9%, 56.3%) reported at least one vulvovaginal symptom. The most common symptom was being dry 35.8% (n/N = 128/358). Ten percent of women (n/N = 38/358) reported ≥ 5 symptoms and 6% of women reported all 7 symptoms in the last week. For women reporting ≥ 1vulvovaginal symptoms, 40.4% (n/N = 74/183) reported emotional impact (Yes to ≥ 1 out of 4 emotional impact subscale items) and 32.8% (n/N = 60/183) reported life style impact (Yes to ≥ 1 out of 5 life impact subscale items) from these symptoms. For sexually active women reporting vulvovaginal symptoms, 75.3% (n/N = 67/89) reported sexual impact (Yes to ≥ 1 out of 4 sexual impact subscale items). Vulvovaginal symptoms were associated with increased co-occurrence of specific pelvic floor disorders, including pelvic organ prolapse (p=0.001), anal incontinence to solid stool (p=0.001), urinary frequency (p=0.02), urgency urinary incontinence (p=0.001), and dysuria (p<0.001).
Vulvovaginal symptoms are common and present in over 50% of postmenopausal women. Sizeable proportions of women with vulvovaginal symptoms report emotional, life style and sexual impact from these symptoms.
Genitourinary syndrome of menopause (GSM) is a new terminology for vulvovaginal atrophy (VVA) formally approved by the International Society for the Study of Women’s Sexual Health and the North American Menopause Society in 2014.(1) VVA includes physical signs of decreased estrogen, symptoms of dryness, burning, irritation, lack of lubrication, discomfort, and pain. This new terminology was proposed because the term atrophy has an unnecessary negative connotation in describing a normal aging process and the unacceptability of the term vagina for use in the media. GSM also encompasses urogenital/lower urinary tract symptoms (LUTS) which include urinary urgency, dysuria and recurrent urinary tract infections.
Three large internet and telephone studies funded by Novo Nordisk, Shionogi, Inc, Pfizer Pharmaceuticals and Wyeth have recently reported the prevalence of GSM symptoms in postmenopausal women the United States, Canada and Europe. (2-5) These works demonstrate that postmenopausal women commonly experience symptoms from GSM. In the REVIVE study of 10,486 U.S. women, 38% of postmenopausal women reported vulvovaginal symptoms in the last 12 months and 59% of women with symptoms reported an impact on their enjoyment of sex.(2) Other themes emerging from these studies were that many women reported a loss of intimacy with their partner or did not seek a new partner due to GSM.(2, 4) Finally, many women believe that vaginal discomfort is a “natural part of growing older” and do not seek medical treatment for these symptoms.(3) While these studies provide both valuable information about the prevalence of VVA in postmenopausal women, impact of VVA symptoms on sexual activity and about treatment seeking behavior and discussions with healthcare providers, less is known about the impact of VVA symptoms on domains of quality of life (QoL) domains, beyond sexual function. Finally, the overlap of symptoms that postmenopausal women experience symptoms from pelvic floor disorders (PFDs) and general menopausal symptoms with GSM is not known.
The development of validated patient reported outcome measures to capture GSM was outlined as a research goal by the Vulvovaginal Atrophy Consensus Panel convened in 2013. In 2010, we began the development process for a valid and internally consistent measure for VVA that would include both symptoms and disease specific quality of life (QoL) impact of VVA symptoms on multiple domains (emotional, lifesyle, and sexual) with multiple items. We first convened an expert panel to develop items, conducted cognitive interviews and finally 3 focus groups to develop the Vulvovaginal Skin Questionnaire (VSQ).(6) The VSQ was then tested for validity and internal consistency using physical exam and other validated questionnaires in 120 postmenopausal women presenting for gynecologic care. In this work, we seek to expand on our previous work by having postmenopausal women complete the VSQ to estimate the prevalence of symptoms and further solidify answer responses.
The primary objective of this study was to use a valid and internally consistent survey instrument, the Vulvovaginal Symptom Questionnaire (VSQ) to estimate the prevalence of vulvovaginal symptoms and examine the emotional, life-impact and sexual-impact of vulvovaginal symptoms in postmenopausal women. Our secondary objective was to determine if responses assessed about degree of bother of vulvovaginal symptoms provided more information than dichotomous responses assessed about the presence or absence of symptoms. Other secondary objectives included a comparison of gynecologic diagnoses, treatments received and hormone therapy (HT) in women with and without vulvovaginal symptoms and an examination of the association between current vulvovaginal symptoms with PFDs and general menopausal symptoms using validated questionnaires.
We conducted a cross-sectional survey. Approval was obtained through the Institutional Review Board (IRB) at Yale University and the Committee for the Protection of Human Subjects (CPHS) serving as the IRB for Dartmouth College.
Our target population was postmenopausal women. Participants were recruited from internal medicine practices in Connecticut and New Hampshire and adult senior centers. Study participation included the anonymous completion of a 15-page written study packet. Study packets were written in English at a 5th grade reading level. The first question of the study packet asked women if they were postmenopausal. Menopause was defined as a woman who “has not had a period or vaginal bleeding for one year or has had both of her ovaries removed.” Women who answered “No” to the question “Have you gone through menopause?” were instructed in writing to stop filling out the study packet.
At the internal medicine practices, a flyer in the waiting room invited interested postmenopausal women to participate in the survey and written copies of the study packet were provided. No study personal was present in primary care waiting rooms. Women then submitted the study packets in locked boxes present in the waiting room. These recruitment procedures precluded the obtainment of a response rate. At the adult senior centers, women were recruited by study personnel.
The study packet included the Vulvovaginal Symptom Questionnaire (VSQ), questions on height, weight, hysterectomy status, vaginal parity, medical conditions that could impact the pelvis (e.g. interstitial cystitis, fibromyalgia, chronic back pain, inflammatory bowel disease, and irritable bowel disease) recent gynecologic treatment and hormone therapy (HT) use, and the following validated questionnaires: The Pelvic Floor Distress Inventory-Short Form 20 (PFDI-20), the Pelvic Floor Impact Questionaire-21 (PFIQ-21), and the Menopause-Specific Quality-of-life Questionnaire. Demographic questions on age, race and ethnicity were obtained from women’s self-report to written answers to questions as a part of the full written study packet.
We have previously published on the validity, reliability, and internal consistency of the VSQ.(6) (see Supplemental Digital Content 1) The VSQ is a self-administered questionnaire with 4 individual subscales: symptoms, emotional-impact, lifestyle-impact and sexual-impact. The first 7 questions of the VSQ comprise the symptom subscale and include questions on itching, burning, hurting, irritation, dryness, discharge and odor in the last week. Women who answered “Yes” to any of the first 7 symptom questions were considered to have vulvovaginal symptoms. Women who answered “No” to all of the first 7 questions were considered to be without vulvovaginal symptoms.
The next two subscales of the VSQ refer to emotional and lifestyle impact of vulvovaginal symptoms in the last week and can be answered by all women, regardless of current sexual activity. After these subscales are completed, the next question, Question 17 asks women if they are currently sexually active with a partner. The answer to this question is not counted as a part of the total score and women who answer no to current sexual activity are instructed to skip the final subscale of sexual impact of vulvovaginal symptoms.
In sexually active women (women who report “Yes” to question 17), the VSQ is a 21-item instrument and scores range from 0 to 20 with increased scores indicating increased vulvovaginal symptom bother. Question 17, which is the question about the report of current sexual activity with a partner, is not counted as a part of the total score. In women who do not report current sexual activity, (“No to question 17), the final 4 questions related to sexual-impact of vulvar symptoms are omitted. The VSQ scores for women who are not sexually active with a partner VSQ range from 0 to 16 with increased scores indicating greater symptom bother.
Our previous work documented that dichotomous answers to VSQ questions (Yes/No) performed as well as a Likert scale recording frequency of bother, e.g. “how often have you been bothered by…,”(6) In this work, we sought to determine if answer responses to degree of bother, .e.g “how much have you been bothered by…,” performed better than dichotomous answers in confirmatory factor analysis modeling. Answer responses to questions first included “Yes” and “No”. The follow-up question on bother was “If you answered yes, how much does this bother you?” with answers of (1) Not at All, (2) Somewhat, (3) Moderately, and (4) Quite a bit.
Women were asked about prior gynecologic treatments sought or received. Women were asked about the timing of their last pelvic examination, if they had seen a gynecologist in the past 2 years, and about current hormone therapy usage. Hormone therapy usage was not asked by product or brand name, but rather by generic hormone name (e.g. estrogen, testosterone). Women were also asked if they had ever had a vulvar biopsy. Women were asked about history of vulvar cancer, vulvar dysplasia (precancerous cells of the vulva), or a vulvar skin condition including lichen sclerosis, lichen planus lichen simplex chronicus, psoriasis, or Bechet’s disease. Women were asked about prior diagnoses of genital herpes and hidradenitis suppurativa. Finally, women were asked about the number of times they were diagnosed with yeast infections and bacteria vaginosis in the last year.
Women were asked to complete 3 additional quality of life questionnaires related to disease-specific gynecologic quality of life in older women. The Menopause Specific Quality of Life questionnaire was used to assess overall menopausal symptom distress. (7) This questionnaire has 4 domains including vasomotor, psychosocial, physical and sexual symptoms. Scores for each domain range between 1 and 8 with increasing scores indicating increased symptom bother. Symptom distress and life-impact for PFDs were assessed utilizing the Pelvic Floor Distress Inventory Short Form-20 (PFDI-20) and the Pelvic Floor Impact Questionaire-7 (PFIQ-7). (8, 9) Scores for both the PFDI-20 and the PFIQ-7 range from 0 to 300 with higher scores indicating more symptom distress or greater life-impact from PFDs. Both questionnaires have sub-scales (range 0 to 100) to measure the contribution of urinary, colorectal-anal, and pelvic organ prolapse symptom distress and life-impact.
Six questions on the PFDI-20 were used to define different female pelvic floor disorders if women answered “Yes” to these questions. Questions 3, 9, 15, 16, 17, and 20 were used define the presence of symptomatic pelvic organ prolapse, fecal incontinence to solid stool, urinary frequency, urgency urinary incontinence, stress urinary incontinence, and dysuria, respectively.
We estimated the occurrence of vulvovaginal symptoms in postmenopausal women to be 38%.(2) Therefore, we calculated needing a sample size of 356 women to determine a point estimate of vulvovaginal symptoms in this population of 38% with a 95% confidence interval of ±5% (e.g. 33%, 43%).
We previously demonstrated the four individual subscales of the VSQ are symptoms, emotional, lifestyle, and sexual impact.(6) Confirmatory factor analysis was performed twice utilizing Mplus software (Muthen & Muthen, Los Angeles, CA).(10-12) First, we performed confirmatory factor analysis modeling on all 358 women (regardless of sexual activity) on the first three subscales of the VSQ (symptoms, emotional, and lifestyle impact). We then performed a second confirmatory factor analysis modeling with all 4 a priori sub-scales of the VSQ using the 166 women who reported current sexual activity and completed the sexual impact subscale. The goodness-of-fit of the two confirmatory factor analysis models were evaluated utilizing the following indices and cutoff levels: comparative fit index (CFI) (≥ 0.95), Tucker Lewis Index (TLI) (≥ 0.95) root mean square error of approximation (RMSEA) (<0.1), and weighted root mean square residual (WRMR) (<0.9).
Women were considered to have vulvovaginal symptoms if they had answered “Yes” to ≥ 1 of the first 7 questions of the VSQ including itching, burning/stinging, hurting, being irritated, being dry, discharge and odor. The occurrence of vulvovaginal symptoms with point estimate and 95% confidence interval (CI) was determined.
Demographic and information on medical conditions were examined in women with no symptoms compared with women with ≥ 1 symptom. We then compared treatments sought and received between women with and without current vulvovaginal symptoms. We compared women with and without vulvovaginal symptoms to other gynecologic symptoms common in postmenopausal women. Specifically, we examined the association of vulvovaginal symptoms with general menopausal symptoms using the Menopause Specific Quality of Life questionnaire and with symptoms of PFDs using the PFDI-20 and the PFIQ-7.
All data analyses were performed using STATA 11.0 (StataCorp, College Station, TX) and Mplus software (Muthén and Muthén, Los Angeles, CA) unless otherwise specified. Fisher’s exact test, student’s t-test and chi-square testing were utilized as appropriate.
Three hundred fifty-eight postmenopausal women completed the survey instrument. Two hundred seventy nine women were recruited from primary care physician’s waiting rooms and 79 women were recruited from senior centers. Participation rates from women approached at senior centers was high. Ninety-six percent (n/N = 79/82) of women approached for participation at senior centers agreed to participate and completed the survey packet. Women participating from primary care practices were younger than women recruited from senior centers (mean age 58.7(±7.2) vs. 73.6 years (±9.3), p<.001). (Table 1) Women participating from primary care practices were more likely to identify their race as white compared with women recruited from senior centers (85.7% (n/N = 239/279) vs. 49.7% (n/N = 39/79), p<.001).
The median age of all women participating in this study was 60 years (range 45 years to 95 years). (Table 2) The majority of women completing this survey identified themselves as white (77.7%). Fifteen percent of women identified themselves as black or African American.
Confirmatory factor analysis modeling was performed with the answer choices from all 358 women for the symptom, emotional and lifestyle impact subscales. When we performed confirmatory factor analysis on the 5-item response scale, the goodness-of-fit of the model was not confirmed. When the answer responses were converted from the continuous scale to a dichotomous response of “Yes/No”, the goodness-of-fit of the confirmatory factor analysis model was confirmed. The indices that confirmed model fit of from the first confirmatory factor analyses on all 358 women for the 3 subscales were: CFI= 0.99, TLI= 0.99 and RMSEA= 0.01. The WRMR= 0.99 was slightly higher than cut-off value of <0.9. The indices that confirmed model fit from the second confirmatory factor analyses on the 168 women who completed the final sexual impact subscale were: CFI =0.99, TLI = 1.00, RMSEA =0.07, and WRMR = 0.83.
One hundred eighty-three post-menopausal women (51.1%; 95% CI 45.9%, 56.3%) reported at least one vulvovaginal symptom in the last week. The most common symptom was being dry 35.8% (n/N = 128/358). (Table 3) Ten percent of women (n/N = 38/358) reported ≥ 5 symptoms and 6% of women reported all 7 symptoms in the last week. (Table 4)
For women reporting ≥ 1 vulvovaginal symptoms, 40.4% (n/N = 74/183) reported emotional impact (Yes to ≥ 1 out of 4 emotional impact subscale items) from these symptoms, and 32.8% (n/N = 60/183) reported lifestyle impact (Yes to ≥ 1 out of 5 lifestyle impact subscale items). One hundred sixty-six women (46.3%) reported currently being sexually active with a partner. Sixty percent (n/N = 101/166) of sexually active women reported current vulvar symptoms and forty-two percent of women (n/N = 82/192) who denied current sexual activity reported symptoms. For sexually active women reporting vulvovaginal symptoms, 76.2% (n/N = 77/101) reported sexual impact (Yes to ≥ 1 out of 4 sexual impact subscale items) from symptoms.
Similar percentage of women with and without symptoms reported that they had seen a gynecologist in the last 2 years (67.5% vs. 66.2%; p=.90). (Table 5) Postmenopausal women with vulvovaginal symptoms were significantly more likely to be on vaginal estrogen therapy compare with women without symptoms (16.9% vs. 3.4%, p<.001). Women with vulvovaginal symptoms were significantly more likely to report a prior vulvar biopsy (p<.001) and more likely to have had one or more yeast infections in the last year (p=.006).
Women with vulvovaginal symptoms had increased menopausal symptoms as measured by the Menopause Specific Quality of Life Questionnaire in the dimensions of psychosocial, physical, and sexual symptoms. The score in the dimension of vasomotor symptoms was not associated with vulvovaginal symptoms (p = .91). Postmenopausal women with vulvovaginal symptoms had significantly increased symptom distress from other pelvic floor disorders including urinary symptoms, colorectal-anal symptoms, and prolapse symptoms compared with women without vulvovaginal symptoms (p<.001 for all) as measured by the PFDI-20 and subscales. (Table 6) Vulvovaginal symptoms were associated with increased co-occurrence of specific pelvic floor disorders, including pelvic organ prolapse (p=0.001), anal incontinence to solid stool (p=0.001), urinary frequency (p=0.02), urgency urinary incontinence (p=0.001), and dysuria (p<0.001). Vulvovaginal symptoms were not associated with a statistically significant co-occurrence of stress urinary incontinence (p=0.1). Postmenopausal women with vulvovaginal symptoms also had significantly higher impact of pelvic floor symptoms on daily life as measured by the PFIQ-7 compared with women without vulvovaginal symptoms. The presence of vulvovaginal symptoms was not associated with increased incontinence impact on the Incontinence Impact Questionnaire subscale of the PFIQ-7 (p=.12).
Among postmenopausal women recruited from adult senior centers and primary care waiting rooms, vulvovaginal symptoms were common, with over 50% of women reporting at least one vulvovaginal symptom within the last week. Over 10% of postmenopausal women reported five or more vulvovaginal symptoms. Current vulvovaginal symptoms impacted on multiple aspects of postmenopausal women’s lives including in the domains of sexual, lifestyle and emotional impact. Additionally, postmenopausal women with vulvovaginal symptoms report increased symptom bother from menopausal symptoms and pelvic floor disorders compared with women without vulvovaginal symptoms.
Our findings regarding the prevalence of vulvovaginal symptoms are consistent with three large online and telephone surveys examining the GSM experienced by postmenopausal women in the United States, Canada, and Europe published in 2012-13.(2-5) The prevalence of GSM in these studies ranged from 38% to 74% with vaginal dryness being the most commonly reported symptom. Extending the work of these prior surveys, we found that current vulvovaginal symptoms impact emotional, lifestyle and sexual aspects of postmenopausal women’s lives. Among women with symptoms, 40% reported an emotional impact and 33% reported a lifestyle impact from these symptoms. Among sexually active postmenopausal women with symptoms, greater than 75% reported a sexual impact from these symptoms. This suggests that GSM may have a sizeable impact on postmenopausal sexual function.
In both our previous work and in this analysis, we found that dichotomous answer responses have consistently performed better in confirmatory factor analysis for the VSQ questionaire than answer responses that seek to elucidate the degree of bother (either how often or how much does this bother you).(6) We propose that by measuring emotional, life and sexual impact in the subscales of the VSQ that perhaps this is serving to reflect true bother or impact of symptoms on postmenopausal women’s lives. In this current work scoring for the VSQ has been clarified, however an important next step in this questionnaire development is to determine the responsiveness of this instrument to change (a.k.a. minimally-important difference) to be able to determine the clinical significance of different treatments of GSM in future trials.
We found among women with current vulvovaginal symptoms, 32.5% have not seen a gynecologist in the last 2 years suggesting an unmet need. In the REVIVE study conducted in the U.S., 56% of post-menopausal women with GSM reported discussing these symptoms with a healthcare provider and just 13% reported their healthcare provider initiated this discussion.(2) We did find women with vulvovaginal symptoms were more likely to receive treatment than women without symptoms. Women with symptoms were more likely to report current vaginal estrogen therapy and prior vulvar biopsy. This finding of increased treatment (vaginal estrogen use and prior vulvar biopsy) with current vulvovaginal symptoms suggests some women are being treated for GSM. However, we found 83.1% of postmenopausal women who reported ≥1 vulvovaginal symptoms were not using vaginal estrogen therapy, the mainstay of treatment for GSM.(13) This lack of treatment among women with current symptoms highlights an area of improvement in gynecologic practice.
We found significant associations with vulvovaginal symptoms and symptom distress from PFDs, including urinary symptoms, and overall menopausal symptoms. The overlap of vulvovaginal symptoms with symptoms of PFDs compliments and advances the concept of GSM including both vulvovaginal symptoms and urinary symptoms such as dysuria and urinary frequency/urgency. Women with ≥ 1 vulvovaginal symptoms (within the last week) were more likely to report symptom distress from PFDs (e.g. urinary incontinence, pelvic organ prolapse, fecal incontinence, and defecatory dysfunction). Women with vulvovaginal symptoms were more likely to report increased menopausal symptom bother from the domains of psychosocial impact, physical symptom and sexual symptom bother. However, women with vulvovaginal symptoms reported similar bother from vasomotor symptoms of menopause compared to women without vulvovaginal symptoms. The association with vulvovaginal symptoms and with other gynecologic disorders, including general menopausal symptoms and specific PFDs, including symptoms of pelvic organ prolapse, anal incontinence, urinary frequency, urgency urinary incontinence, and dysuria, not only supports the new expanded terminology of GSM, but the concept that these symptoms encompass not just the vaginal and vulva, but also the lower urinary tract.
Women were recruited from senior centers and internal medicine practices and were not seeking gynecologic specialty care. However given the anonymous nature of our subject recruitment; we do not have a response rate. We hypothesize that the true prevalence of vulvovaginal skin symptoms in all postmenopausal women is slightly lower than this current study population as asymptomatic women may have elected not to complete the study questionnaires displayed in the waiting rooms. However, women were not told that this study was studying vulvovaginal symptoms. Recruitment flyers simply stated we were studying “common symptoms in post-menopausal women” and did not focus on GSM. Information obtained on emotional, lifestyle and sexual impact of vulvovaginal symptoms is cross-sectional; therefore causality cannot be determined. Additionally, our analyses are univariate and there may be confounding that is not adjusted for. We are limited in our information about treatment seeking behavior and treatments received which was self-reported and not verified in the medical record. Information on hormonal therapy was asked in general questions about specific hormones (e.g. estrogen, progesterone, and testosterone) and not brand name products. We do not have specific information about non-hormonal lubricants or selective estrogen receptor modulators that have been demonstrated to impact GSM. We also do not have information about the timing of treatment in relationship to symptoms. Finally, we do not know how many women or their health care providers initiated conversations about GSM. These are all areas for future works.
We have found that vulvovaginal symptoms are experienced by approximately 50% of postmenopausal women and that women with vulvovaginal symptoms demonstrate emotional, lifestyle impact and sexual impact from these symptoms. We have further demonstrated that women with vulvovaginal symptoms report increased gynecologic treatment (vaginal estrogen therapy and prior vulvar biopsies). However, we have also demonstrated an unmet need with over 30% of post-menopausal women with vulvovaginal symptoms reporting not seeing gynecologist in over 2 years and over 80% of postmenopausal women with vulvovaginal symptoms are not using vaginal estrogen therapy. These findings suggest significant areas for future interventions. Finally, we have demonstrated significant associations of vulvovaginal symptoms with other gynecologic symptoms experience by older women including PFDs, urinary symptoms, and general menopausal symptom bother supporting the new expanded definition of GSM.
This work was supported in part by a grant from the Claude D. Pepper Older Americans Independence Center at Yale University School of Medicine (P30AG021342 NIH/NIA) and an award from the National Institutes on Aging (R03AG042363-01).
No reprints available
DISCLOSURE: No conflicts of interest identified.
Supplemental Digital Content
Supplemental Digital Content 1: The Vulvovaginal Symptom Questionnaire .doc