Urinary incontinence is a common problem in middle-aged and older women associated with decreased quality of life.1 Prevalence estimates in midlife women range from approximately 5% for severe to 60% for mild incontinence.2,3 Because prevalent incontinence has been associated with postmenopausal status by clinicians and in epidemiologic studies,4 it is often considered a symptom attributable to the change from the reproductive estrogen levels of premenopause to the lower estrogen levels of postmenopause as occurs across the menopausal transition.5
The menopause transition is a series of stages of variable length from premenopause, early perimenopause, and late perimenopause to postmenopause, defined by changes in menstrual and hormonal patterns. Results from cross-sectional epidemiologic studies show an increase in prevalence of incontinence among women between the ages of 45 and 55 years, an age range that coincides with the menopause transition.2 Some studies have found no association between the menopausal transition and the risk for developing incontinence,6 but this association remains uncertain.
In a previous set of analyses, we showed that the worsening of incontinence was not associated with menopausal transition stage in women with incontinence.7 Our main objective in the present analyses was to estimate whether menopause transition stages were independently associated with the development of any stress or urge urinary incontinence symptoms in women who were continent at baseline. We also considered whether other modifiable factors that change over time, such as changes in weight and medical conditions, affected the development of incontinence.
PATIENTS AND METHODS
We analyzed data from 1,529 community-dwelling women enrolled in the Study of Women’s Health Across the Nation (SWAN) who reported no incontinence at baseline and followed them through six annual follow-up visits (1995–2002) (Fig. 1). SWAN is a multicenter, multiethnic, prospective study of the menopause transition8 funded by the National Institute of Aging. Seven clinical sites (Boston, MA; Chicago, IL; the Detroit area, MI; Los Angeles, CA; Newark, NJ; Pittsburgh, PA; and Oakland, CA) recruited a total cohort of 3,302 women. Eligibility criteria for entry into the SWAN cohort were age 42–52 years and self-identification as one of five racial or ethnic groups (African American, Hispanic, Chinese, Japanese, white). Exclusion criteria included inability to speak English, Spanish, Japanese, or Cantonese, no menstrual period in greater than 3 months before enrollment, hysterectomy or bilateral oophorectomy before enrollment, and current pregnancy, lactation, or hormone use. The institutional review boards at all sites approved SWAN, and all women gave informed consent to participate.
A self-administered questionnaire assessed incontinence at baseline and at each follow-up visit. Based on response to the question: “In the past year/since your last study visit, have you ever leaked even a small amount of urine involuntarily?” we classified frequency of incontinence as “almost daily/daily” (daily), “several days per week” (weekly), “less than one day per week” (monthly), “less than once a month,” or “none.” We defined any incontinence as incontinence occurring at least monthly. We considered incontinence occurring less than once a month as clinically insignificant and subject to higher misclassification and thus combined this category with “no incontinence.” We categorized type of incontinence as “stress” if participants reported leakage with “coughing, laughing, sneezing, jogging, jumping, with physical activity, or picking up an object from the floor” or as “urge” if participants reported leakage “when you have the urge to void and can’t reach the toilet fast enough.” We defined “any” incontinence as either stress or urge symptoms. Women who reported new onset incontinence at any of the six annual follow-up visits were compared with women who did not develop incontinence over the same time period.
Study of Women’s Health Across the Nation classified menopausal status annually from menstrual bleeding patterns. Premenopause was less than 3 months of amenorrhea and no menstrual irregularities in the previous year; early perimenopause was less than 3 months of amenorrhea and some menstrual irregularities in previous year; late perimenopause was 3 to 11 months of amenorrhea; and postmenopause as 12 consecutive months of amenorrhea. Women who used hormone therapy (oral contraceptives or systemic estrogen and/or progestin) before the last menstrual period were considered to have an “unclassifiable” menopausal status at the time of use. Similarly, we could not classify the menopausal stage of women who underwent hysterectomy because data on whether one or both ovaries remained in situ could not be confirmed. To assess whether change in menopausal status was associated with the development of incontinence, we created a variable with four mutually exclusive categories comparing the status in the previous year to the current year, ie, change in menopausal stage (eg, from premenopause to early perimenopause); started hormones (from any stage to hormone use); or stopped hormones (from hormone use to none) compared with no change in stage or hormone use.
We calculated body mass index (BMI) as weight in kilograms/(height in meters)2 based on measurements taken annually by certified staff who used calibrated scales and a stadiometer. Socioeconomic status was approximated by level of difficulty paying for basics (food, heat, and shelter). Interviewers obtained self-reported medical histories, smoking history, and medication use. Each year, SWAN used the same questions from the Center for Epidemiological Studies–Depression scale9 (for characterizing our cohort, we defined depressive symptoms as a score of 16 or above), the Medical Outcomes Study Social Support Survey,10 the Life Stressors and Social Resources Inventory,11 and the Psychiatric Epidemiology Research Interview.12 Study of Women’s Health Across the Nation measured anxiety symptoms by a summed score of days in the past 2 weeks in which certain symptoms were experienced (grouchiness or irritability, feeling tense or nervous, pounding or racing heart, feeling fear for no reason); for characterizing our cohort, we defined anxiety symptoms by a score of 4 or more.13 At year 1 only, SWAN combined responses to questions assessing sensitivity to physical sensations into a Symptom Sensitivity Scale.14
Our fixed or baseline covariates included baseline age, race or ethnicity, BMI, diabetes, hypertension, self-reported diagnosis of leiomyomas, parity, marital status, socioeconomic status, education, social support, general health, and symptom sensitivity. Time-dependent covariates were menopausal status, hormone use before the final menstrual period, new self-reported diabetes and hypertension, and smoking status (never, ever, current). We created variables to represent change in certain characteristics by subtracting values in the current year from values in the previous year: weight change (per pound), waist-to-hip ratio change (per 0.1 units), anxiety and depressive symptoms score change (per one unit), change in the number of stressful life events (per one event), change in social support scores (per one unit), and overall health status change on a five-point scale.
Dropouts were those deceased, who discontinued the study voluntarily, or who could not be contacted after missing two or more visits at the sixth annual visit. When a woman was missing data on frequency and type of incontinence from one or two visits, we imputed values as follows: If the missing value occurred at year 6, we imputed by using the value at the previous visit. If women reported no incontinence in the years prior and subsequent to a missing response, we assumed no incontinence in those missing years. If a woman was missing incontinence data in the 1 to 2 years prior to a first report of incontinence, we randomly assigned her missing values to either no incontinence or the frequency or type of incontinence in that subsequent year. We imputed incontinence frequency for 324 women (13.5%) and type for 18 women (0.2%). When weight was missing for one or two visits, the we imputed values for 88 visits (1.2%) as the mean between the two known values. In the same way, we imputed waist circumference values for 138 visits (1.8% of all visits). For all other independent variables, we dropped missing data from the analysis. We excluded New Jersey year 6 data from our models because that site had disproportionate and systematic loss of the Hispanic and white participants during that year.
We compared proportions and means of each variable for women who were continent at baseline and who remained in or dropped out of the study using the t and χ2 tests. For our survival analysis, we used discrete proportional hazards models.15 First, in our main models, we evaluated whether menopausal stage at the annual visit concurrent with the first report of incontinence and other time-dependent factors were associated with the development of monthly or more any incontinence compared with no development of any incontinence. Here we included a variable to account for whether a woman had advanced to another menopausal stage or had started or stopped hormones. We created similar separate models for stress and urge incontinence. For stress incontinence, our comparison group was those women who had no development of stress incontinence, and for urge incontinence it was those women who had no development of urge incontinence. Second, we evaluated menopausal stage in the year prior to the first report of incontinence, controlling for the same covariates. Finally, we examined whether concurrent menopausal stage and change in status was associated with the development of more frequent incontinence by modeling weekly or more incontinence compared with no development of incontinence. The candidate covariates described above were chosen based on the literature, a priori hypotheses, or were associated with the outcome in univariable analysis at P<.10. We used SAS 9.1 (SAS Institute Inc., Cary, NC).
We evaluated proportional hazards assumptions by plotting log(–log[survival]) functions for various groupings of the data based on the time independent covariates and estimated survival functions for each group using the Kaplan-Meier estimates. Because the proportional hazards held regardless of the covariate groupings, the assumption seemed to be adequate for all the fixed and time-dependent covariates. We tested model stability by running models with and without imputed data and respectively forcing in menopausal status, change in status, and age without significant changes in the point estimates of the other variables. Correlation matrices showed no significant colinearity among our independent variables, including between menopausal status and age. We chose our final models based on the lowest Akaike Information Criterion, which indicates the best fit for discrete proportional hazards models.
In the subset of SWAN participants who did not report incontinence at baseline, we compared relevant baseline characteristics of the 1,529 who remained in the study (the follow-up cohort) with the 279 who we classified as lost to follow-up (dropout cohort) as of the sixth follow-up (Table 1). Women who dropped out were more likely to be Hispanic, were less likely to have gone to college, had a lower socioeconomic status, had a higher BMI, reported poorer health, and were more likely to smoke.
During the 6-year study period, 41.3% of the women transitioned from premenopause or early perimenopause to postmenopause; 6.4% had undergone hysterectomy with or without oophorectomy, and 6.7% had used hormones. By year 6, 28.7% of the women had gained greater than 5% of their baseline body weight, and only 7.8% had lost the equivalent weight. Of the women who were continent at baseline, 739 reported new onset monthly or more incontinence over the 6 years of follow-up (8.7 per 100 per year), whereas 108 reported new onset weekly or more incontinence (2.3 per 100 per year). Overall, the incidence of incontinence in woman-years decreased after perimenopause (Fig. 2).
Being in late perimenopause was associated with an increased hazard of developing any incontinence and urge incontinence during the 6 years of follow-up compared with premenopause (Table 2). For any incontinence, being in early perimenopause also increased the hazard. Although not statistically significant, the estimated hazard ratios (HRs) for developing stress and urge incontinence in early perimenopause were similar in magnitude and direction to the estimated HR for any incontinence in this menopausal stage. Whether menopausal status changed or women started or stopped using hormones was not associated with the development of incontinence in that year and did not stay in our models.
Because we could not determine precisely when incontinence developed between two annual visits, we also examined menopausal stage in the year before the first report of incontinence. We found the estimated HRs to be similar or larger for any incontinence and stress incontinence compared with our models for menopausal stage in the concurrent year (Table 3). We also found that being postmenopausal significantly decreased the hazard for the development of any incontinence and stress incontinence in the subsequent year. The HRs for urge incontinence were in a similar direction for each menopausal status, but were not statistically significant. Again, whether menopausal status changed or women started or stopped using hormones was not associated with the development of incontinence and did not stay in our models.
The development of any incontinence was associated with a higher baseline BMI, weight gain, and an increase in anxiety symptoms (Table 2). A higher baseline BMI and weight gain were associated with new onset stress incontinence, whereas an increase in anxiety symptoms were associated with urge incontinence. The estimated HR for the association between weight gain and urge incontinence was similar to that for any incontinence and stress incontinence but did not meet our definition of statistical significance. Neither the number of live births nor the route of delivery (vaginal compared with cesarean delivery) were associated with incontinence development (data not shown).
We also evaluated factors associated with the development of more frequent (weekly or more) incontinence (Table 4). We observed no association between menopausal status and the development of weekly or more incontinence. Although our numbers in this analysis were small (n=108), the estimated HRs were closer to 1.00. Some factors were the same as in our main models for any monthly or more incontinence (baseline BMI, weight gain and an increase in anxiety symptoms), but the HRs were larger. Development of diabetes was a unique factor associated with approximately a 50% increased hazard for developing weekly or more incontinence.
We found that early and late perimenopause stages were associated with an increase, and that postmenopause was associated with a decrease in the hazard for developing monthly or more frequent incontinence. However, the development of weekly or more frequent incontinence (leaking several times per week or more) was not associated with any stage of the menopausal transition. Additionally, advancing from one menopausal stage to the next was not associated with the development of incontinence. This suggests that the risk of developing only mild incontinence (leaking less than several times per week) was affected by the menopausal transition. Although not statistically significant, both stress and urge incontinence showed similar patterns of positive and negative association.
A biologic basis may explain the association between menopausal stage and the reporting of new onset incontinence. Some women report that their incontinence is affected by menstrual cycle phase, with most women reporting increases in incontinence in the luteal phase when both estradiol and progesterone levels are elevated.16 Exogenous hormone use in postmenopausal women has a weak negative effect on both stress and urge urinary incontinence.17 Sharp declines in estradiol levels have been associated with a lower risk of urinary incontinence.18 It is possible that for women in the perimenopause, the increased frequency of anovulatory cycles and the associated relatively prolonged elevated levels or peaks of estrogen in a subset of women19 increases the likelihood of developing infrequent incontinence, whereas the lower or declining estrogen levels of postmenopause decreases that risk.
In addition to hormonal or biologic effects of progression through the menopausal transition on incontinence, women may change reporting behavior over the menopause transition. Women transitioning through menopause have other symptoms such as hot flushes, nighttime awakening, and changes in vaginal bleeding, especially in the perimenopause, and increased annoyance with these symptoms may heighten reporting of incontinence that in and of itself was less bothersome in other menopausal stages. Increases in vaginal discharge could be mistaken for new onset incontinence during the early perimenopause. Additionally, women in the perimenopause and postmenopause may change unmeasured behaviors that could initially increase and then decrease the chance of developing incontinence. For example, women may change diet and exercise habits that unmask incontinence symptoms as they transition from premenopause to perimenopause.
A few other associations with the development of incontinence in midlife women are worth noting. Although cross-sectional studies have described a relationship between depression20 or anxiety21 and incontinence, our longitudinal analysis demonstrates that the development of anxiety symptoms was strongly associated with incident any monthly or more incontinence and urge monthly or more incontinence. The strength of this association doubled with the development of more frequent (weekly or more) incontinence. Higher anxiety could increase the likelihood of reporting of incontinence, or a physiologic relationship may exist. For example, the sympathetic nervous system stimulation associated with anxiety could affect the sensitivity and activity of the nerves to the lower urinary tract or affect the resting tension, fatigability, or function of the levator ani muscles of the pelvic floor that help maintain continence.22 Weight gain among midlife women was common—almost 30% of the women in SWAN gained 5% or more of their body weight over the first 6 years—and was associated with an increased risk for developing any and stress incontinence, likely related to increased intraabdominal pressures.23
The diagnosis of diabetes in the previous year increased the likelihood of reporting new onset weekly or more frequent incontinence. Diabetes is consistently one of the strongest risk factors for prevalent urinary incontinence.24 How diabetes increases risk is not known, but potential mechanisms include diabetic-associated neuropathy, ischemia, and polyuria due to glucosuria.
Some factors associated with incident incontinence in this survival analysis with time-dependent covariates are similar to those we have identified with incontinence in two previous SWAN analyses. For example, we found that baseline BMI was associated with both prevalent and incident incontinence in our logistic regression models of baseline factors only,25 and baseline BMI and weight gain were both associated with worsening incontinence in our generalized estimating equation longitudinal models.7 The greater intraabdominal pressure of high BMIs and weight gain can theoretically overload the continence mechanism over time, leading to the development of incontinence in continent women and worsening incontinence in incontinent women. Other factors, such as baseline self-reported health status associated with incident incontinence in our logistic regression models25 were not associated with incident incontinence in this survival analysis. The advantage of longitudinal analysis is that specific factors affecting health status such as new onset medical problems and changes in mental health symptoms proximal to the first report of incontinence can be modeled and may explain this discrepancy.
Our study had a number of limitations. With prospective cohort studies, dealing with loss to follow-up and missing data are challenging. Although SWAN’s retention rate of 86.1% over 6 years is very good, the differences between those who remained in the study at 6 years and those who dropped out potentially introduces bias and limits the generalizability of our results. A higher proportion of women who reported no incontinence at baseline dropped out of the study, possibly inflating our incidence rates and reducing the size of our comparison group. Our definition of incident incontinence may also have overestimated our incidence rates in the first few annual follow-up visits. Some women who reported no incontinence at baseline may have had incontinence in previous years that was not troubling them at the time of study. Because the population of continent women at risk declined over the 6 years of this study, women most at risk for incontinence may have been culled from the study in the earlier years. Finally, the incontinence questions used in this study were not from validated questionnaires.
The study also had important strengths. Seven years of longitudinal data on incontinence in more than 2,800 racially or ethnically diverse, community-based, midlife women is remarkable. It permits generation of incidence during the menopausal transition, information that has been noted to be limited by the National Institutes of Health state-of-the-science conference on urinary incontinence.26 The same incontinence questions were asked on an annual basis; these questions, which allowed classification of incontinence symptoms by clinical type (stress and urge), were similar to validated questions that have been developed since the inception of SWAN and those that have been used widely in other epidemiologic studies assessing incontinence.27 The sensitivity and specificity of self-reported incontinence by type is estimated at 71–85% and 60–79%, respectively, in validated questionnaires.28 Finally, these findings are consistent with the results of our previous study evaluating factors associated with worsening symptoms in incontinent women.7
In the popular health media, urinary incontinence is often listed as a symptom of menopause, but the findings from SWAN indicate that this is inaccurate. Although we found an association between the development of incontinence and the perimenopausal stages, the incontinence was limited to mild symptoms reported to occur less than weekly. Taken with the findings of our previous study showing that incontinence symptoms improve in the perimenopause,7 this suggests that incontinence during this stage is likely to be transient and self-limited. Midlife women should be counseled that their risk of developing incontinence in the early postmenopause is less than in the years before and during the menopausal transition. From a public health standpoint, clinicians and women should focus on the other factors consistently shown to be risk factors for both the development and worsening of incontinence: being overweight, weight gain, and the development of diabetes. Because a number of these factors are modifiable, determining whether healthy life changes and treating associated medical problems can prevent the development of incontinence is a priority.
1. Mitteness LS. The management of urinary incontinence by community-living elderly. Gerontologist 1987;27:185–93.
2. Hunskaar S, Arnold EP, Burgio K, Diokno AC, Herzog AR, Mallett VT. Epidemiology and natural history of urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct 2000;11:301–19.
3. Sampselle CM, Harlow SD, Skurnick J, Brubaker L, Bondarenko I. Urinary incontinence predictors and life impact in ethnically diverse perimenopausal women. Obstet Gynecol 2002;100:1230–8.
4. Thom DH, Brown JS. Reproductive and hormonal risk factors for urinary incontinence in later life: a review of the clinical and epidemiologic literature. J Am Geriatr Soc 1998;46:1411–7.
5. Bachmann G. Urogenital ageing: an old problem newly recognized. Maturitas 1995;22:S1–S5.
6. Sherburn M, Guthrie JR, Dudley EC, O’Connell HE, Dennerstein L. Is incontinence associated with menopause? Obstet Gynecol 2001;98:628–33.
7. Waetjen LE, Feng WY, Ye J, Johnson WO, Greendale GA, Sampselle CM, et al. Factors associated with worsening and improving urinary incontinence across the menopausal transition. Obstet Gynecol 2008;111:667–77.
8. Sowers M, Crawford S, Sternfeld B, Morganstein D, Gold EB, Greendale GA, et al. SWAN: a multicenter, multiethnic, community-based cohort study of women and the menopausal transition. In: Lobo RA, Kelsey JL, Marcus R, editors. Menopause: biology and pathobiology. San Diego (CA): Academic Press; 2000. p. 175–88.
9. Radloff L. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385–401.
10. Sherbourne CD, Stewart AL. The MOS social support survey. Soc Sci Med 1991;32:705–14.
11. Moos RH, Fenn CB, Billings AG, Moos BS. Assessing life stressors and social resources: applications to alcoholic patients. J Subst Abuse 1988;1:135–52.
12. Dohrenwend B, Dohrenwend B. Life stress and illness: formulation of the issues. New York (NY): Prodist; 1981.
13. Gold EB, Block G, Crawford S, Lachance L, FitzGerald G, Miracle H, et al. Lifestyle and demographic factors in relation to vasomotor symptoms: baseline results from the Study of Women’s Health Across the Nation. Am J Epidemiol 2004;159:1189–99.
14. Barsky AJ, Goodson JD, Lane RS, Cleary PD. The amplification of somatic symptoms. Psychosom Med 1988;50:510–9.
15. Diggle P, Heagerty P, Liang K, Zeger S. Analysis of longitudinal data. 2nd ed. New York (NY): Oxford University Press; 2002.
16. Hextall A, Bidmead J, Cardozo L, Hooper R. The impact of the menstrual cycle on urinary symptoms and the results of urodynamic investigation. BJOG 2001;108:1193–6.
17. 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.
18. Gopal M, Sammel MD, Arya LA, Freeman EW, Lin H, Gracia C. Association of change in estradiol to lower urinary tract symptoms during the menopausal transition. Obstet Gynecol 2008;112:1045–52.
19. Santoro N, Brown JR, Adel T, Skurnick JH. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab 1996;81:1495–501.
20. Vigod SN, Stewart DE. Major depression in female urinary incontinence. Psychosomatics. 2006;47:147–51.
21. Bogner HR, Gallo JJ, Swartz KL, Ford DE. Anxiety disorders and disability secondary to urinary incontinence among adults over age 50. Int J Psychiatry Med 2002;32:141–54.
22. Deffieux X, Hubeaux K, Porcher R, Ismael SS, Raibaut P, Amarenco G. Pelvic floor muscle activity during coughing: altered pattern in women with stress urinary incontinence. Urology 2007;70:443–7.
23. Richter HE, Creasman JM, Myers DL, Wheeler TL, Burgio KL, Subak LL. Urodynamic characterization of obese women with urinary incontinence undergoing a weight loss program: the Program to Reduce Incontinence by Diet and Exercise (PRIDE) trial. Int Urogynecol J Pelvic Floor Dysfunct 2008;19:1653–8.
24. Ebbesen MH, Hannestad YS, Midthjell K, Hunskaar S. Diabetes and urinary incontinence—prevalence data from Norway. Acta Obstet Gynecol Scand 2007 Sep 4:1–7. [Epub ahead of print].
25. Waetjen LE, Liao S, Johnson WO, Sampselle CM, Sternfield B, Harlow SD, et al. Factors associated with prevalent and incident urinary incontinence in a cohort of midlife women: a longitudinal analysis of data: study of women’s health across the nation. Am J Epidemiol 2007;165:309–18.
26. NIH state-of-the-science conference statement on prevention of fecal and urinary incontinence in adults. NIH Consens State Sci Statements 2007;24:1–37
27. Brown JS, Grady D, Ouslander JG, Herzog AR, Varner RE, Posner SF. Prevalence of urinary incontinence and associated risk factors in postmenopausal women. Heart & Estrogen/Progestin Replacement Study (HERS) Research Group. Obstet Gynecol 1999;94:66–70.
28. Brown JS, Bradley CS, Subak LL, Richter HE, Kraus SR, Brubaker L, et al. The sensitivity and specificity of a simple test to distinguish between urge and stress urinary incontinence. Ann Intern Med 2006;144:715–23.