Despite the effect of female pelvic floor disorders on women’s quality of life,1–4 the significant financial cost of its management and treatment,5 and the increased occurrence of these conditions, pelvic floor disorders remain underreported and undertreated, making assessment of the prevalence of these disorders difficult. Underreporting has been attributed to women being too embarrassed to discuss the issues with their physicians, and the belief that incontinence is a normal part of aging.1,6 Most of the estimates for pelvic floor disorders are derived from the incidence of surgery for these disorders,7,8 or from clinic-based samples, which may overestimate the prevalence of these conditions. Little is known about women who are managed conservatively or those who have not sought medical attention for these conditions. Estimates of prevalence for incontinence vary widely, from 10% to 58% for urinary incontinence (UI), and from less than 1% to 39% for anal incontinence.9,10 Although pelvic organ prolapse (POP) remains one of the most common indications for hysterectomy, few data exist on its prevalence. There is a paucity of information about the co-occurrence of pelvic floor disorders, such as the association between fecal and urinary incontinence or between POP and incontinence, with reported prevalence estimates ranging from 1.3% to 9.4% for double incontinence.11,12
The absence of solid epidemiologic data on the prevalence and co-occurrence of female pelvic floor disorders has hampered efforts to improve the care of women with these disorders and to identify preventive strategies for these conditions. The Kaiser Permanente Southern California Continence Associated Risk Epidemiology Study used a validated instrument, the Epidemiology of Prolapse and Incontinence Questionnaire (EPIQ), to assess a broad spectrum of pelvic floor disorders in a large, racially and ethnically diverse population across a wide age range from a community-dwelling population. We assessed the prevalence of four pelvic floor disorders and examined the co-occurrence of these conditions within age categories.
MATERIALS AND METHODS
Kaiser Permanente, the setting for this study, is a large prepaid managed health care plan that serves more than three million residents in southern California. Kaiser Permanente members are similar to persons in the seven-county geographic area served by the health plan in southern California. Although the membership does not tend to include the very indigent or the very wealthy, the median income and the racial and ethnic composition of the membership are similar to the southern California population.
Development and pilot testing of the EPIQ as well as survey methods have been described elsewhere.13 Briefly, the EPIQ was developed and validated to assess the presence or absence of stress urinary incontinence (SUI), overactive bladder (OAB), POP, and anal incontinence. Each condition was criterion validated by comparing responses on EPIQ to physical examination findings and a report of symptoms bothersome enough to seek treatment. The criteria of symptoms bothersome enough to seek treatment was established a priori by an expert panel and validated using focus groups to estimate the potential demand for treatment of these conditions and the burden of the disorders in community-dwelling women. The conditions of OAB included urgency and frequency, with or without incontinence, and anal incontinence included flatal and solid and/or liquid incontinence. Positive and negative predictive values for the detection of specific conditions are 76% and 97% for POP, 88% and 87% for SUI, 77% and 90% for OAB, and 61% and 91% for anal incontinence, respectively.14
After approval by the Kaiser Permanente Southern California Institutional Review Board, a sample of 12,200 women, just over twice the desired final sample size, was generated by enumerating all 950,000 female Kaiser Permanente Southern California members in April 2004 between the ages of 25 and 84 years and randomly selecting 3,050 women within four age strata; 25–39 years, 40–54 years, 55–69 years, and 70–84 years from all women who had an address on file with the health plan. Surveys in English and Spanish were mailed to these women with a cover letter introducing the survey, small incentive, postcard to opt out or request additional information, followed by a second survey mailing and a reminder telephone call and a third survey mailing to women in the youngest age stratum. If an additional survey was requested at the time of the reminder call, it was sent with an additional cover letter outlining the importance of the study and was followed up with an additional reminder telephone call if it was not returned. Consent was implied by completion of the survey. Data were collected from April 2004 through January 2005.
Each pelvic floor disorder was categorized dichotomously as “present” or “absent” based on responses to each symptom domain and degree of bother with these symptoms.13 Women for whom we did not have information to assess the presence or absence or the disorder were excluded from any analyses for that outcome. Age was modeled as a categorical variable in four 15-year age strata. Potentially confounding variables were obtained by self-report from the EPIQ. Body mass index (BMI) was dichotomized into obese (30 kg/m2 or more) and not obese (less than 30 kg/m2). Menopause status and hormone use were concatenated into one variable with four categories: premenopausal, menopausal with no hormone use, menopausal with past hormone use, and menopausal with current hormone use. Menopause was also analyzed separately as a dichotomous variable without respect to hormone use for subanalyses of women in the perimenopausal age groups. Smoking was categorized as never smoked, past smoker or current smoker. Chronic lifting was defined as lifting of more than 9 kg (excluding children) regularly for more than one year. Caffeine intake was defined as more than one cup of caffeinated beverage per day. Medical comorbidities (neurologic disease, lung disease or asthma, diabetes, history of depression), hysterectomy, menopause status, and hormone exposure were assessed based on affirmative responses to questions on the survey.
Number of vaginal deliveries was characterized based on pregnancy history and method of delivery (vaginal or cesarean). A combined variable categorized women as nulliparous (no vaginal deliveries more than 2,000 g), cesarean delivery only (no vaginal deliveries more than 2,000 g), one vaginal birth, two vaginal births, and three or more vaginal births of infants weighing more than 2,000 g. For this study, a birth weight of less than 2,000 g was considered to have insignificant effect on the pelvic floor.
To assess the co-occurrence of pelvic floor disorders, we restricted our analyses to a subset of women in whom we could assess the presence or absence of all four pelvic floor disorders. We present overall prevalence and 95% confidence interval (CI) of each pelvic floor disorder in this subsample of women and the total number of women with the condition by age group. Then we present the proportion and 95% CI of women who have the specific pelvic floor disorders exclusively and the proportion that has it in combination with other pelvic floor disorders assessed in this study as a fraction of the total sample with the condition. Categories are not mutually exclusive, but rather assess the most commonly observed combinations of pelvic floor disorders.
Power calculations and sample size were determined from published estimates of the prevalence of pelvic floor disorders9,10 to address the primary aim of the study: to assess the associations between parity, mode of delivery, and the presence of common pelvic floor disorders and to test the hypothesis that vaginal delivery is associated with an increased risk of pelvic floor disorders compared with cesarean delivery.14 Power analysis assumed the prevalence of SUI and OAB to be 20% and that of anal incontinence and POP to be 6%. We assumed equal prevalence of pelvic floor disorders in the nulliparous and the cesarean groups, with the prevalence in the vaginally parous group being 1.5 times that of the other two. We estimated that 15% of the women in the study population were nulliparous, 15% only had cesarean deliveries, and 70% were vaginally parous. Using these assumptions, 5,400 respondents were needed to achieve 80% power to detect a 50% difference with a significance of .05. Assuming a 45% response rate, a total of 3,050 women in each 15-year age category were selected for inclusion in the sample.
We present the prevalence estimates and 95% CI for each of the pelvic floor disorders stratified by the four age categories and for the entire sample. We assessed the differences between age groups using the χ2 test and between mean ages using the Student t test. Significance was evaluated using a P value of less than .05.
Multiple logistic regression analysis was used to calculate the adjusted odds ratios (ORs) and 95% CI for age groups while controlling for the potential and known risk factors for each pelvic floor disorders. Once all of the variables were entered into the model, we removed covariates that were no longer significant in the multivariable model except for age category, race/ethnicity, number of vaginal births, menopause/hormone status, hysterectomy, and obesity, which remained in every model. Fully adjusted models including each significant variable and partially adjusted models including only the strongest associated variables were performed. Statistical analyses were performed with SAS 9.1 (SAS Institute, Cary, NC).
Of the 4,458 (37%) surveys returned, 140 were in Spanish and 4,318 in English. After excluding 289 surveys due to insufficient information to assess pregnancy history and mode of delivery and 66 surveys due to insufficient information to assess any of the four pelvic floor disorders, survey responses from 4,103 women were available for these analyses. Of these women, 764 were 25–39 years, 981 were 40–54 years, 1,187 were 55–69 years, and 1,171 were 70–84 years of age. About two thirds of the women were married or living with a partner and less than 30% had no vaginal deliveries with birth weights more than 2,000 g. Almost three quarters had some college education, one quarter were obese, and two thirds were postmenopausal (Tables 1 and 2).
Nonresponders were significantly younger than responders (53±17 years compared with 57±16 years, P<.001). Racial/ethnic distribution of women in our analytic sample did not differ significantly from all others surveyed (P=.26). When we compared women in the analytic sample to all other women originally surveyed, 10% of the women in the sample and 11% of the remaining women had diabetes (P<.05).15 Other medical, historical and demographic data for Nonresponders are not available for comparison.
The overall prevalence (95% CI) of each pelvic floor disorder for the 4,103 women was 15% (14.1–16.3%) for SUI, 13% (12.3–14.4%) for OAB, 6% (5.7–7.2%) for POP, and 25% (23.7–26.4%) for anal incontinence. The prevalence of any one or more pelvic floor disorders was 37% (35.8–38.9%). The overall prevalence of each condition varied significantly by age category (Fig. 1). The unadjusted prevalence of OAB and anal incontinence significantly increased with age (P<.01); whereas SUI significantly increased with age, peaked in the 55–69-year age group, and then fell in the oldest age group (P<.01). The prevalence of POP remained relatively stable throughout all the age groups (P=.10).
After adjusting for the common significant confounding factors, which included birth experience, hormone/menopause status, hysterectomy, and obesity, age group was no longer significantly associated with the prevalence of any individual pelvic floor disorder with the exception of POP in the 55–69-year-old age group and anal incontinence in the 40–54-year-old age group (Table 3). Fully adjusted multivariable models for each disorder included additional significant variables (history of smoking, lifting, caffeine use, frequent urinary tract infections, diabetes, depression, lung disease, neurologic disease, and diuretic use) and resulted in age group no longer being significantly associated with any pelvic floor disorder (data not shown). Overall trends in these models did not differ from the partially adjusted models presented in Table 3. To control for any covariation between menopause status and age and to ensure that menopause was not artificially confounding the age category variable in the models, subanalysis of the 40–54-year-old and 55–69-year-old age groups was performed to assess the association between menopause status and each pelvic floor disorder within these two perimenopausal age groups. Being menopausal remained significantly positively related to each pelvic floor disorder except POP (SUI 1.56, 1.16–2.10; OAB 1.79, 1.26–2.54; POP 1.40, 0.86–2.28, anal incontinence 1.65, 1.28–2.13).
The covariates most consistently associated with increasing odds of each pelvic floor disorder were being obese, increasing vaginal parity, hormone usage, and history of hysterectomy. Obesity was consistently positively associated with each pelvic floor disorder and demonstrated the highest odds for SUI (2.55) and OAB (2.73). Vaginal parity was positively associated with each pelvic floor disorder with the greatest effect on POP, where the odds were 3- to 3.5-fold higher in women with more than one vaginal delivery compared with nulliparous women. The highest odds for each pelvic floor disorder were seen in women with three or more vaginal deliveries. Past or current hormone use among menopausal women was also positively associated with each pelvic floor disorder when compared with premenopausal women. Postmenopausal women who never used hormone did not differ from premenopausal women. Hysterectomy was associated with all conditions except anal incontinence. Race/ethnicity was only associated with decreased anal incontinence in black women.
Among the subset of 3,799 women (93% of the total sample) for whom we could assess the presence or absence of all four conditions, we observed that younger women (25–39 years) with pelvic floor disorders were most likely to have only one condition and not be affected by other conditions (Table 4). Co-occurrence of pelvic floor disorders was highest in the oldest age groups for all combinations except for SUI and anal incontinence, and OAB and anal incontinence, which were most common in the 55–69-year-old women, and POP and SUI, which were most common in the 40–54-year-old age group. Overall, the most common combination of pelvic floor disorders seen in this population was SUI and anal incontinence at 9% (95% CI 7.9–9.7%) (333/3,799). The least common combination of conditions was having all four pelvic floor disorders, at 1% (0.9–1.6%) (47/3,799).
Of all women with any lower urinary tract symptoms (SUI, OAB, or SUI and OAB), 67% (63.2–69.8%) had an additional pelvic floor disorder. A total of 81% (78.0–84.0%) of women with SUI symptoms, 80% (76.7–83.3%) of women with OAB symptoms, and 48% (44.5–51.0%) of women with anal incontinence had additional pelvic floor disorders. For the small proportion of women with POP, 69% (62.3–74.5%) were identified as having an additional pelvic floor disorder. Twenty percent (18.6–21.2%) of women in the sample had some lower urinary tract symptoms, either SUI or OAB. Most women with OAB reported leakage of urine (96%, 94.2–97.7%) (484/504) with an overall prevalence of 12% (11.7–13.9%) for OAB wet. The prevalence of mixed SUI and OAB wet (mixed urinary incontinence) was 8% (7.5–9.2%) (316/3,799). Of those women with anal incontinence, 38% (34.8–41.1%) (358/945) had flatal incontinence only, leaving a prevalence of fecal incontinence of 16% (14.3–16.6%). Mixed urinary and fecal incontinence (double incontinence) prevalence was 4% (3.2–4.5%) (145/3,799).
To date, the prevalence and co-occurrence of pelvic floor disorders has not been well established. The co-occurrence of pelvic floor disorders in our study was high, with 50% to 80% of women with one of the four pelvic floor disorders having at least one other pelvic floor disorder. The few studies that have addressed this issue focus on mixed urinary and fecal incontinence, also known as double incontinence. Findings from a community-based survey yielded a 9% prevalence for combined fecal and urinary incontinence,11 which is higher than our estimate of 4%. The discrepancy between these two estimates may be related to differences in definitions and severity of the conditions between the two studies. A study of 2,000 community-dwelling women on the prevalence of urinary and fecal incontinence and POP symptoms found an age-adjusted 1.8-fold increased odds of anal incontinence in women with urinary incontinence, an unadjusted 5- to 6-fold increased odds of fecal incontinence in women with urinary incontinence, and a nearly 4-fold increase in urinary and fecal incontinence in women with pelvic heaviness (a surrogate for POP).16
The prevalence of any one or more pelvic floor disorder in this population was also high at 37% and increased significantly from the youngest age group to the oldest age group before adjustment for other factors. These trends are consistent with previous studies.17,18 Our estimates of 13% for OAB and 15% for SUI are somewhat lower than the prevalence estimates of 18%, 25%, and 28% for urinary incontinence reported in other studies.19–21 These studies reported increasing prevalence with age. These prevalence estimates may be higher than ours because the definitions used for these studies were “any incontinence” without respect to degree of bother. Similar to our findings, OAB prevalence was reported to be 17% in U.S. women, with increasing prevalence by age.22
The prevalence of fecal incontinence has been reported to be 12% in women aged older than 65 years, 7% in community dwelling women 30 years to 90 years, 15% in women older than 50 years, and an age-adjusted prevalence of 14% in community-dwelling women, with positive association between increasing age and prevalence of pelvic floor disorders.11,23–25 These prevalence estimates seem somewhat lower than those we report. However, our study uses a broader definition of anal incontinence, which includes leakage of gas as well as solid and liquid stool. If we exclude women with leakage of gas only, our fecal incontinence estimates are comparable (16%). A recent study of 2,000 women with mean age of 55.6±8.6 years reported a 6% estimate of POP based on physical examination26 and, consistent with our findings, found no association between prevalence of POP and age.
We found that after adjusting for confounding variables, age group alone no longer represented an important contributing factor to the prevalence of any single pelvic floor disorder. Rather, common comorbid conditions such as obesity, hormone therapy, prior hysterectomy, and vaginal parity seem to represent more important correlates for pelvic floor disorders than age alone. These associated conditions have been reported by the Kaiser Permanente Southern California Continence Associated Risk Epidemiology Study and other studies.14–16,26–29
Despite considerable effort to increase our response rate, particularly among younger health plan members, it still remained lower than anticipated. Younger members were hardest to reach; the likelihood of having the survey returned as undeliverable by the post office decreased with increasing age, from 11% of 25 to 39 year olds to 3% of 70 to 84 year olds. Because members must update their contact information to schedule an appointment and older members, on average, have more frequent appointments than younger members, this is not unexpected. Because older women are more likely to have pelvic floor disorders, our estimates of the prevalence of each condition will overestimate the prevalence in our population. To address this limitation, we focused on prevalence estimates by age category after presenting overall estimates. Our prevalence estimates are fairly comparable to the published literature because many studies only survey older women by design. The smaller than expected sample size and the low prevalence of POP limits our power to detect differences by age category for this outcome if they existed. Also, the overall prevalence estimates for each condition are limited by the sensitivity and specificity of our survey, which is lowest for the identification of anal incontinence. However, these characteristics are not available for other instruments used in published articles, and thus, comparison is not feasible.
As with most studies evaluating the independent relationship of age and any condition in women, it is difficult to separate the individual contribution. We attempted to control for this by conducting a subanalysis of women in the perimenopausal age groups and found that menopause remained a significant contributor to the presence of each pelvic floor disorder. Thus, we do not feel that menopause is merely acting as a surrogate for age group. Although respondents were slightly less likely to have diabetes than nonrespondents, we found that obesity, either alone or with diabetes, demonstrated a greater association with SUI, OAB, and anal incontinence than did having diabetes alone.15 Although we do not have information on the height and weight of nonrespondents, our overall prevalence of obesity was consistent with what we would have anticipated given national estimates.
The major strengths of this study are that we were able to present prevalence estimates for four common pelvic floor disorders and describe the co-occurrence of these conditions in one sample using consistent methodology. To date, there have been very limited data on the co-occurrence of these conditions in a community-dwelling sample. While the differential response by age category and the potential for unmeasured differences in respondents and nonrespondents may have affected our prevalence estimates, this should not effect the associations between the four pelvic floor disorders under study and the covariates most consistently associated with increasing odds of each pelvic floor disorder nor our presentation of the co-occurrence of these conditions.
Pelvic floor disorders are common, their overall prevalence increases with age, and they often occur in combination with other pelvic floor disorders. However, although pelvic floor disorders may be more common with increasing age, our data suggest that they may not be an inevitable part of the ageing process. Women should not be resigned to suffer from pelvic floor disorders as a normal part of the ageing process; rather, they should focus on modifiable risk factors for prevention such as weight loss and maintenance, and seek treatment for all the conditions when they occur. In this population, 67% to 81% of women with urinary symptoms and more than 50% with POP or anal incontinence reported at least one other pelvic floor disorder. Physicians seeing women seeking care for a female pelvic floor disorder should be aware of this high co-occurrence of these conditions and should inquire about symptoms of other pelvic floor disorders.
1. Brocklehurst JC. Urinary incontinence in the community—analysis of a MORI poll. BMJ 1993;306:832–4.
2. Grimby A, Milsom I, Molander U, Wiklund I, Ekelund P. The influence of urinary incontinence on the quality of life of elderly women. Age Ageing 1993;22:82–9.
3. Grimby A, Svanborg A. Morbidity and health-related quality of life among ambulant elderly citizens. Aging (Milano) 1997;9:356–64.
4. Wyman JF, Harkins SW, Fantl JA. Psychosocial impact of urinary incontinence in the community-dwelling population. J Am Geriatr Soc 1990;38:282–8.
5. Wagner TH, Hu TW. Economic costs of urinary incontinence in 1995. Urology 1998;51:355–61.
6. Reymert J, Hunskaar S. Why do only a minority of perimenopausal women with urinary incontinence consult a doctor? Scand J Prim Health Care 1994;12:180–3.
7. Allard P, Rochette L. The descriptive epidemiology of hysterectomy, Province of Quebec, 1981–1988. Ann Epidemiol 1991;1:541–9.
8. Lepine LA, Hillis SD, Marchbanks PA, Koonin LM, Morrow B, Kieke BA, et al. Hysterectomy surveillance—United States, 1980–1993. MMWR CDC Surveill Summ 1997;46:1–15.
9. Bump RC, Norton PA. Epidemiology and natural history of pelvic floor dysfunction. Obstet Gynecol Clin North Am 1998;25:723–46.
10. Mallett VT, Bump RC. The epidemiology of female pelvic floor dysfunction. Curr Opin Obstet Gynecol 1994;6:308–12.
11. Roberts RO, Jacobsen SJ, Reilly WT, Pemberton JH, Lieber MM, Talley NJ. Prevalence of combined fecal and urinary incontinence: a community-based study. J Am Geriatr Soc 1999;47:837–41.
12. Thomas TM, Egan M, Walgrove A, Meade TW. The prevalence of faecal and double incontinence. Community Med 1984;6:216–20.
13. Lukacz ES, Lawrence JM, Buckwalter JG, Burchette RJ, Nager CW, Luber KM. Epidemiology of prolapse and incontinence questionnaire: validation of a new epidemiologic survey. Int Urogynecol J Pelvic Floor Dysfunct 2005;16:272–84.
14. Lukacz ES, Lawrence JM, Contreras R, Nager CW, Luber KM. Parity, mode of delivery, and pelvic floor disorders. Obstet Gynecol 2006;107:1253–60.
15. Lawrence JM, Lukacz ES, Liu IL, Nager CW, Luber KM. Pelvic floor disorders, diabetes, and obesity in women: findings from the Kaiser Permanente Continence Associated Risk Epidemiology Study. Diabetes Care 2007;30:2536–41.
16. Uustal Fornell E, Wingren G, Kjolhede P. Factors associated with pelvic floor dysfunction with emphasis on urinary and fecal incontinence and genital prolapse: an epidemiological study. Acta Obstet Gynecol Scand 2004;83:383–9.
17. Luber KM, Boero S, Choe JY. The demographics of pelvic floor disorders: current observations and future projections. Am J Obstet Gynecol 2001;184:1496–501.
18. MacLennan AH, Taylor AW, Wilson DH, Wilson D. The prevalence of pelvic floor disorders and their relationship to gender, age, parity and mode of delivery. BJOG 2000;107:1460–70.
19. Peyrat L, Haillot O, Bruyere F, Boutin JM, Bertrand P, Lanson Y. Prevalence and risk factors of urinary incontinence in young and middle-aged women. BJU Int 2002;89:61–6.
20. Rortveit G, Hannestad YS, Daltveit AK, Hunskaar S. Age- and type-dependent effects of parity on urinary incontinence: the Norwegian EPINCONT study. Obstet Gynecol 2001;98:1004–10.
21. Melville JL, Katon W, Delaney K, Newton K. Urinary incontinence in US women: a population-based study. Arch Intern Med 2005;165:537–42.
22. Stewart WF, Van Rooyen JB, Cundiff GW, Abrams P, Herzog AR, Corey R, et al. Prevalence and burden of overactive bladder in the United States. World J Urol 2003;20:327–36.
23. Maaita M, Bhaumik J, Davies AE. Sexual function after using tension-free vaginal tape for the surgical treatment of genuine stress incontinence. BJU Int 2002;90:540–3.
24. Melville JL, Fan MY, Newton K, Fenner D. Fecal incontinence in US women: a population-based study. Am J Obstet Gynecol 2005;193:2071–6.
25. Bharucha AE, Zinsmeister AR, Locke GR, Seide BM, McKeon K, Schleck CD, et al. Risk factors for fecal incontinence: a population-based study in women. Am J Gastroenterol 2006;101:1305–12.
26. Rortveit G, Brown JS, Thom DH, Van Den Eeden SK, Creasman JM, Subak LL. Symptomatic pelvic organ prolapse: prevalence and risk factors in a population-based, racially diverse cohort. Obstet Gynecol 2007;109:1396–403.
27. Brown JS, Sawaya G, Thom DH, Grady D. Hysterectomy and urinary incontinence: a systematic review. Lancet 2000;356:535–9.
28. Hendrix SL, Cochrane BB, Nygaard IE, Handa VL, Barnabei VM, Iglesia C, et al. Effects of estrogen with and without progestin on urinary incontinence. JAMA 2005;293:935–48.
© 2008 The American College of Obstetricians and Gynecologists
29. Hendrix SL, Clark A, Nygaard I, Aragaki A, Barnabei V, McTiernan A. Pelvic organ prolapse in the Women’s Health Initiative: gravity and gravidity. Am J Obstet Gynecol 2002;186:1160–6.