Skip Navigation


American Journal of Epidemiology Advance Access originally published online on November 28, 2006
American Journal of Epidemiology 2007 165(3):309-318; doi:10.1093/aje/kwk018
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
165/3/309    most recent
kwk018v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Waetjen, L. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Waetjen, L. E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2006 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A.

ORIGINAL CONTRIBUTIONS

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

L. Elaine Waetjen1, Shanmei Liao2, Wesley O. Johnson3, Carolyn M. Sampselle4, Barbara Sternfield5, Siobán D. Harlow6, Ellen B. Gold7 and for the Study of Women's Health Across the Nation

1 Department of Obstetrics and Gynecology, University of California, Davis, CA
2 Department of Statistics, University of California, Davis, CA
3 Department of Statistics, University of California, Irvine, CA
4 School of Nursing, University of Michigan, Ann Arbor, MI
5 Kaiser Permanente Department of Research, Oakland, CA
6 Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI
7 Department of Public Health Sciences, University of California, Davis, CA

Correspondence to Dr. L. Elaine Waetjen, Department of Obstetrics and Gynecology, 4860 Y Street, Suite 2500, Sacramento, CA 95817 (e-mail: lewaetjen{at}ucdavis.edu).

Received for publication February 1, 2006. Accepted for publication July 5, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
To compare the characteristics of and baseline factors associated with prevalent and incident urinary incontinence in a diverse cohort of midlife women, the authors analyzed the baseline and first five annual follow-up visits of the Study of Women's Health Across the Nation (SWAN), 1995–2001. From responses to annual questionnaires, the authors defined prevalent incontinence as at least monthly incontinence reported at baseline and incident incontinence as at least monthly incontinence first reported over follow-up. They used multiple logistic regression for their comparison. The mean age of their cohort at baseline was 45.8 (standard deviation: 2.7) years. Prevalent incontinence was 46.7%, and the average incidence was 11.1% per year. Most women reported stress, but a higher proportion developed urge incontinence (15.9% vs. 7.6% at baseline). African Americans (29.5%) and Hispanics (27.5%) had the lowest prevalence of incontinence; African Americans (11.6%) and Caucasians (13.4%) had the highest average annual incidence. Parity, diabetes, fibroids, and poor social support were associated with prevalent incontinence, while high body mass index, high symptom sensitivity, and poor health were associated with incident incontinence. In midlife women, incident incontinence is mild with different characteristics and baseline risk factors; overweight women have a higher risk of developing incontinence.

cohort studies; incidence; middle aged; prevalence; urinary incontinence; women

Abbreviations: CI, confidence interval; OR, odds ratio; SWAN, Study of Women's Health Across the Nation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The prevalence of urinary incontinence in women peaks in midlife, between the ages of 45 and 55 years (1), yet the epidemiology of incontinence in this age group is understudied relative to research in women aged 60 years or more. Incontinence that develops in midlife is not well characterized, and little is known about factors that may predispose women to the different clinical types of incontinence during this time.

Most large, community-based, epidemiologic studies of female incontinence have been cross-sectional or retrospective and conducted predominantly in older women of European descent (1–5). Additionally, most have assessed risk factors for prevalent incontinence in these populations with uncertainty as to the duration of incontinence. Yet, incontinence that develops in midlife may have unique characteristics and risk factors.

The purpose of this study was to compare prevalent stress, urge, and mixed incontinence with incontinence that developed over 5 years of follow-up in a racially and ethnically diverse cohort of midlife women. We used data from the Study of Women's Health Across the Nation (SWAN), a community-based, prospective cohort study of women from five different racial/ethnic groups, to determine whether risk factors for prevalent incontinence such as parity, high body mass index, and diabetes were also associated with incident incontinence. We also examined whether factors associated with prevalent and incident stress, urge, and mixed incontinence varied by race/ethnicity.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
This analysis used data from the baseline and first five annual follow-up visits from the prospective cohort of SWAN, 1995–2001. SWAN is a multicenter, multiethnic, and multidisciplinary prospective cohort study of the menopausal transition. Details of the study's methodology have been published previously (6). At seven clinical sites (Boston, Massachusetts; Chicago, Illinois; the Detroit area, Michigan; Los Angeles, California; Newark, New Jersey; Pittsburgh, Pennsylvania; and Oakland, California), 16,065 community-based women, aged 40–55 years, were identified by random digit dialing, snowball, and/or list-based sampling and interviewed for a cross-sectional screening survey. From this large sample, each of the sites recruited about 450 women to include a Caucasian group and one designated minority group (African American at four sites and Chinese, Japanese, and Hispanic at one site each) for a total longitudinal cohort of 3,302 women. Eligibility criteria for the SWAN longitudinal cohort were age 42–52 years and self-identification as one of the five racial/ethnic groups to be studied. The exclusion criteria were inability to speak English, Spanish, Japanese, or Cantonese; no menstrual period in more than 3 months before enrollment; hysterectomy and/or bilateral oophorectomy prior to the onset of the study; and current use of oral contraceptives, estrogens, progestins, or luteinizing hormone agonists. The institutional review boards at each clinical site and the coordinating center approved the study protocol, and all participants gave written, informed consent. For this analysis, to compare factors associated with prevalent and incident incontinence in the same cohort, we excluded 599 women who dropped out of the study before year 5. We considered women to have dropped out of the study when they were missing 3 or more consecutive years of outcome data.

Covariates
At baseline, participants completed both interview-administered and self-administered questionnaires so that demographic, medical, lifestyle, quality of life, and other information could be obtained. Race and ethnicity were self-defined as Black or African American, non-Hispanic Caucasian, Chinese or Chinese American, Japanese or Japanese American, and Hispanic (including Central American, Cuban or Cuban American, Dominican, Mexican or Mexican American, Puerto Rican, South American, Spanish, or other Hispanic). Socioeconomic status was assessed by income and level of difficulty in paying for basics. Interviewers obtained medical history, obstetric history, gynecologic history including self-reported diagnosis of fibroids, smoking history and medication use, educational level attained, and general health status. SWAN used an adaptation of the 36-Item Short-Form Health Survey (SF-36) (7) to assess quality of life, the Center for Epidemiologic Studies Depression (CES-D) Scale (8) to evaluate depressive symptoms, the Baeke questionnaire (9) to assess physical activity, and the Medical Outcomes Study (MOS) Social Support Survey (10) and Life Stressors and Social Resources Inventory (LISRES) (11) to estimate social support. Finally, at year 1, women responded to a series of questions assessing their sensitivity to physical sensations (heat, cold, noise, hunger); the responses were combined into a Symptom Sensitivity Scale (12).

Each woman underwent weight, height, and waist circumference measurements by certified staff that used calibrated scales and a stadiometer. We calculated body mass index (weight (kg)/(height (m)2) and measured waist circumference in centimeters.

Outcomes
The self-administered questionnaire assessed incontinence at baseline and at each annual follow-up visit. Women were asked the question: "In the past year (or since your last study visit), have you ever leaked even a small amount of urine involuntarily?" The frequency of incontinence was recorded 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 classified the responses indicating weekly or daily incontinence as "frequent" and the responses indicating monthly as "occasional." We defined "any incontinence" as at least monthly incontinence. Because we considered incontinence that occurred less than once a month not to be clinically significant with a higher misclassification rate, we combined this category with none as "no incontinence" or no regular incontinence.

We categorized type of incontinence as "stress" if participants reported at least monthly leakage with "coughing, laughing, sneezing, jogging, jumping, with physical activity, or picking up an object from the floor" or as "urge" if participants reported at least monthly leakage "when you have the urge to void and can't reach the toilet fast enough." Affirmative responses to both circumstances of at least monthly incontinence in the same visit were categorized as "mixed." We classified incontinence that did not fit these three types as "other."

We defined prevalent incontinence as incontinence occurring at least monthly that was reported by women at baseline. Prevalence was calculated as the number of women reporting incontinence (by frequency and type) at baseline divided by the total baseline cohort. Women who had no incontinence at baseline but reported at least monthly incontinence at any of the five annual follow-up visits were considered to have incident incontinence at their first report. We calculated cumulative incidence as the the number of women reporting new-onset incontinence in the 5 years of follow-up divided by the number of women at risk during those 5 years. For the average annual incidence, we then divided this by five. Finally, we defined our reference group narrowly: Only women who did not report incontinence occurring at least monthly at baseline and over all 5 years of follow-up were considered to have no incontinence. For 169 women who were missing incontinence data at one of the first five visits, we imputed values for incontinence frequency and type by randomly assigning the individual participant's reported frequency and type from either the previous or subsequent year to the missing year.

Statistical analysis
After univariate exploration of our variables, we performed bivariate analyses. Variables associated with incontinence in the literature or with p values of 0.10 or less in our data were entered into multivariable analyses by use of a forward stepwise process. Additionally, because race, parity, and body mass index (3, 13–15) have all been associated with incontinence, we forced these variables into most of our models.

Because we wanted to compare baseline factors associated with prevalent and first reported incident incontinence within the same cohort, we used multiple logistic regression for our analyses of the dichotomous outcomes: any incontinence versus none; frequent (daily or weekly) incontinence versus none. Our primary models to evaluate factors associated with at least monthly incontinence by type used the following outcomes, each compared with none: urge, stress, or mixed incontinence. When sample size permitted, we developed separate models for prevalent and incident incontinence. We also developed separate models for each racial/ethnic group to study whether factors associated with incontinence varied by race/ethnicity. Because of small numbers in the Chinese, Japanese, and Hispanic cohorts, prevalent and incident factors associated with the individual types of incontinence could not be evaluated separately in these three models. We assessed model fit with the Hosmer-Lemenshow goodness-of-fit test.

To explore the robustness of our results, we developed other models to compare with our primary ones reported here. For urge and stress symptoms, we developed secondary models comparing the following: urge and mixed incontinence (any urge symptoms) versus stress incontinence and none (no urge symptoms), and stress and mixed incontinence (any stress symptoms) versus urge incontinence and none (no stress symptoms). Additionally, because the distributions of body mass index were significantly different among racial/ethnic groups, we explored a number of methods to handle the differences in distributions. First, we entered body mass index by each race/ethnicity as interaction terms into our models. Second, we ran our models categorizing body mass index by quantiles specific to each racial/ethnic group to evaluate body mass index independent of race. Finally, we removed data from all women in the top 5 percent of the body mass index distributions. Although we observed no large differences in our point estimates in each of these methods, model fit by the Hosmer-Lemenshow goodness-of-fit test was generally better when all women remained in the analysis. For all our analyses, we used SAS, version 9.1, software (SAS Institute, Inc., Cary, North Carolina).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Study sample
Of the 3,302 women in the cohort at baseline, 599 (18.1 percent) either dropped out of SWAN or had not completed the incontinence questions at baseline and in all 5 years of follow-up. Table 1 displays the characteristics of the baseline participants and compares the women who remained in the study at year 5 with those women who dropped out of the study or were missing incontinence data. As expected, we found a number of differences between the follow-up and drop-out cohorts. Women who remained in the study were more likely to be Caucasian, to have gone to college, and to be financially secure. Our follow-up cohort was also thinner and had lower parity. Importantly, we also found differences in incontinence reporting. Women who remained in the study were more likely to report any incontinence and more likely to report frequent incontinence. However, the distribution of incontinence type did not vary between follow-up and drop-out cohorts.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Baseline characteristics and differences between women who followed up and dropped out* of the Study of Women's Health Across the Nation over 5 years between 1995 and 2001

 
Prevalent and incident incontinence
The prevalence and cumulative incidence by frequency and type of incontinence for all racial/ethnic groups are displayed in tables 2 and 3. At least monthly incontinence prevalence was 46.7 percent overall and 15.3 percent for incontinence occurring several days per week or more. The average 1-year incidence of at least monthly incontinence was 11.1 percent per year, but only 1.2 percent per year was reported to be frequent. African-American and Hispanic women had the lowest prevalence of incontinence, while African-American (11.6 percent) and Caucasian (13.4 percent) women had the highest average annual incidence. Therefore, African-American women had the highest proportional difference in reporting incident compared with prevalent incontinence at baseline. Incontinence that developed over the 5 years was less frequent (5.8 vs. 15.3 percent had frequent incontinence) and less bothersome (3.2 vs. 4.8 on a Likert scale). Although stress incontinence was the most frequent type of both prevalent and incident incontinence, women of all racial/ethnic groups reported twice as much urge incontinence at follow-up than at baseline (15.9 vs. 7.6 percent at baseline). For both prevalent and incident incontinence, Japanese and Hispanic women were the most likely to report stress incontinence, while African-American women were the most likely to report urge incontinence.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Prevalence* of at least monthly incontinence reported at baseline by racial/ethnic group, Study of Women's Health Across the Nation, 1995–2001

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Cumulative rates* of first reported at least monthly incident incontinence by racial/ethnic group over 5 years, Study of Women's Health Across the Nation, 1995–2001

 
Factors associated with prevalent and incident incontinence
In our multivariable analyses, we found some differences in the factors associated with any or frequent prevalent incontinence compared with any or frequent incident incontinence (table 4). For example, compared with Caucasians, African-American, Chinese, Japanese, and Hispanic women had lower odds of reporting frequent prevalent incontinence, while we observed no significant trend in the reporting of any or frequent incident incontinence by racial/ethnic group. Parity, diabetes, uterine fibroids, depressive symptoms, and poor social support were significantly associated with prevalent incontinence but not incident incontinence. We found a significant interaction between education and both Chinese and Japanese ethnicity. A lower level of education was associated with less prevalent incontinence (for Chinese: odds ratio (OR) = 0.33, 95 percent confidence interval (CI): 0.17, 0.69; for Japanese: OR = 0.48, 95 percent CI: 0.25, 0.97). Being either a current or past smoker was associated with increased odds of incident incontinence in African-American women but not in other racial/ethnic groups. Factors associated with frequent prevalent or incident incontinence are similar to those for any incontinence, but the odds ratios tended to be higher for frequent incontinence.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Adjusted* odds ratios for any and frequent prevalent and incident urinary incontinence, Study of Women's Health Across the Nation, 1995–2001

 
Diabetes was the strongest risk factor that we identified for prevalent incontinence. The estimated probability that a parous woman with a normal body mass index had incontinence at baseline increased with diabetes for each racial/ethnic group, but it was highest in Caucasians (without diabetes: p = 0.51; with diabetes: p = 0.68) and lowest in Hispanics (without diabetes: p = 0.17; wth diabetes: p = 0.29). Although the odds of having incontinence at baseline increased with waist circumference, a high body mass index was associated with developing incontinence during follow-up. The estimated probability of a parous woman's developing incident incontinence with a body mass index of 30 or more was 0.41 for Caucasians, 0.50 for African Americans, 0.39 for Chinese, 0.31 for Japanese, and 0.30 for Hispanics, while for a normal body mass index of 19–24.9, it was 0.33 for Caucasians, 0.42 for African Americans, 0.31 for Chinese, 0.24 for Japanese, and 0.23 for Hispanics.

Factors associated with stress, urge, and mixed prevalent and incident incontinence
Factors associated with prevalent and incident incontinence differed when analyzed by type of incontinence (table 5). Although parity was associated with increased odds of prevalent stress incontinence, it was not associated with incident stress incontinence. Diabetes and waist circumference also were associated with increased odds of prevalent stress incontinence, but body mass index was associated with incident stress incontinence. Prevalent urge incontinence was negatively associated with higher education (college or more compared with high school or less) and had a strong positive association with diabetes. Incident urge incontinence was associated with increasing waist circumference. Prevalent mixed incontinence was associated with predictors of both stress and urge incontinence, as well as with unique factors. In particular, low social support, depressive symptoms, and a high symptom sensitivity score were associated with prevalent mixed incontinence symptoms. Incident mixed incontinence was associated with diabetes, a high body mass index, and a high symptom sensitivity score. Race/ethnicity categories could not be forced into this mixed incontinence model because of poor model fit.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Adjusted* odds ratios for at least monthly prevalent and incident stress, urge, and mixed urinary incontinence, Study of Women's Health Across the Nation, 1995–2001

 
The proportional increase in African-American women reporting incident incontinence can largely be explained in our multivariable models by their higher odds of developing urge incontinence, not stress incontinence. Chinese, Japanese, and Hispanic women had less stress, urge, or mixed incontinence compared with Caucasian women; although not statistically significant, they tended to have less incident incontinence of all types as well (table 5).

Factors associated with incontinence by race
We had sufficient numbers to evaluate factors associated with incontinence type by race/ethnicity in women with prevalent incontinence for our two largest groups, Caucasian and African-American women. Interestingly, parity was significantly associated with prevalent stress and mixed incontinence in Caucasian women (for stress: OR = 1.85, 95 percent CI: 1.18, 2.92; for mixed: OR = 2.51, 95 percent CI: 1.35, 4.69) but not in African-American women (for stress: OR = 0.46, 95 percent CI: 0.11, 1.91; for mixed: OR = 0.51, 95 percent CI: 0.11, 2.41). Although low social support was important only in the reporting of mixed incontinence for Caucasian women (OR = 2.49, 95 percent CI: 1.41, 4.38), African-American women with low social support were more likely to report stress and mixed incontinence (for stress: OR = 2.32, 95 percent CI: 1.21, 4.45; for mixed: OR = 2.35, 95 percent CI: 1.19, 4.66). Finally, self-reported diagnosis of uterine fibroids was associated with urge incontinence in African-American women only (OR = 1.95, 95 percent CI: 1.07, 3.54), while poor health status was associated with stress and urge incontinence in Caucasian women only (for stress: OR = 4.49, 95 percent CI: 1.03, 19.64; for urge: OR = 9.27, 95 percent CI: 1.72, 49.97).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
This study is among the first to examine the prevalence and incidence of urinary incontinence in a racially/ethnically diverse, community-based sample of midlife women followed longitudinally over 5 years. Nearly half of the women reported having at least monthly incontinence at baseline, while half of those who did not have incontinence at baseline reported developing at least monthly incontinence over the 5 years of observation. The characteristics of and factors associated with prevalent and incident incontinence differed. Although Caucasians reported the highest incontinence prevalence and incidence, African-American women had the highest proportional difference between reporting incident and prevalent incontinence; they also had the highest proportion of urge incontinence.

Incident urinary incontinence that developed in this racially/ethnically diverse longitudinal cohort of midlife women was infrequent, occurring less than weekly, compared with the prevalent incontinence reported at baseline. Although stress incontinence symptoms were more frequently reported, the proportion of women reporting urge incontinence symptoms was greater in women with incident compared with prevalent incontinence.

Incontinence described as occurring less than weekly is responsible for the midlife prevalence peak of incontinence in previous epidemiologic studies (1). Whether we captured an early stage of incontinence that will worsen over time or whether incontinence that develops in midlife is more likely to remain mild or even to resolve is not known at this time and will be the subject of future investigations in SWAN. The prevalence of urge incontinence is low in premenopausal women (1) and increases with age (3, 15, 16); the timing of changes in that distribution has been unclear. Our study suggests that this rate begins to change in midlife.

Some factors associated with incident incontinence appeared to be unique. Most important, while parity was associated with increased odds of prevalent stress and mixed incontinence, it was not an important factor in the development of any type of incident incontinence over the 5 years of follow-up. This novel observation suggests that the impact of parity on incontinence presents in the reproductive years, while incontinence that develops in midlife is not related to child-bearing, but rather may be related to other factors to be explored in future studies, such as weight gain and change in hormone status. We also found that, although a higher waist circumference was associated with prevalent incontinence at baseline, higher body mass index was associated with incident incontinence. While this may be, in part, an artifact of our model, it may also relate to the increased proportion of urge symptoms in incident incontinence. We found that a higher waist circumference was associated with stress incontinence; perhaps greater central obesity increases stress incontinence symptoms because of increased intraabdominal pressure, but it has no impact on urge incontinence symptoms for which simply being overweight is a risk factor.

Some factors associated with prevalent stress, urge, and mixed incontinence differed between the two largest racial/ethnic groups: Caucasians and African Americans. In particular, incontinence of any type in African Americans was not associated with parity, while parity was an important factor in Caucasian women. We found that urge incontinence in African Americans was associated with self-reported fibroids, while incontinent Caucasians, but not African Americans, were more likely to report a fair-to-poor health status.

African Americans have been reported to have a lower prevalence of incontinence (14, 15, 17), but they report a higher proportion of urge incontinence (18, 19) compared with other racial/ethnic groups. We found that a higher proportion of African-American midlife women reported incident than prevalent incontinence. One possible explanation for this finding is the association between urge incontinence and fibroids, which are more prevalent in African-American women in this age group (20, 21). Because urinary urgency is thought to be a symptom of fibroids, the association between fibroids and urge incontinence may simply reflect a higher rate of diagnostic studies in African-American women, rather than a direct effect of fibroids. On the other hand, the space-occupying effects of fibroids on the bladder and urethra may indeed be a cause of urge incontinence. Another explanation for the higher reporting of incident incontinence is that the African-American women in our sample had the highest average body mass index compared with the other racial/ethnic groups.

Our study also has important limitations. First, as expected, African-American and Hispanic women, women with less education, and women with a lower socioeconomic status were more likely to have dropped out of SWAN. Women who had higher body mass indexes and who reported poorer health, factors associated with incontinence, were also more likely to be lost to follow-up. Most important, women who reported any and frequent incontinence were more likely to have stayed in SWAN. These differences, though relatively small in absolute numbers, limit the generalizability of our study. In particular, we have likely overestimated the prevalence and incidence of incontinence in midlife women. However, our estimates of factors associated with prevalent incontinence are similar to those previously reported in an analysis of the entire baseline cohort in SWAN (13). Second, we defined prevalent incontinence as incontinence that was reported to be at least once per month and present at baseline. We do not have information to determine the duration of this incontinence, which may have been present for only a few months or for many years. Third, some of our analyses were limited by small numbers. For example, because of small sample sizes, we were unable to evaluate factors associated with the different types of incontinence in our Chinese, Japanese, and Hispanic cohorts. That we identified fewer factors associated with incident incontinence may be, in part, explained by smaller numbers in this group. Fourth, because our incontinence questions have not been validated in diverse populations, reporting bias for incontinence based on varying sociocultural interpretation of these questions may have occurred. For example, our finding that poor social support was associated with urge incontinence in African Americans may reflect psychosocial circumstances that lead to the higher reporting of urge incontinence in this group. Finally, although of greatest interest is the effect of the menopausal transition and other time-varying factors on the natural history of incontinence in midlife, the first 5 years of SWAN do not provide sufficient numbers of women in each stage of the menopausal transition to allow a robust analysis of this effect.

Nonetheless, SWAN has provided a unique opportunity to describe and evaluate incident incontinence in a racially/ethnically diverse, community-based cohort of women over time and had a number of advantages. This longitudinal cohort has had good retention (80 percent) over the first 5 years of follow-up. The same incontinence questions were asked on an annual basis; these questions, which allowed classification of incontinence by clinical type (stress, urge, and mixed), were similar to those that have been used widely in other epidemiologic studies and that have been associated with different risk factors (3, 15). Finally, although self-report reflects the symptoms rather than the diagnosis of stress and urge incontinence, the experience of incontinence is of more direct clinical and public health importance than the presence or absence of urodynamic abnormalities.

Our study evaluating the baseline factors associated with incontinence in the cohort of midlife women provides novel and clinically relevant epidemiologic information about urinary incontinence that develops in this age group. First reported incident incontinence is infrequent, with a higher proportion of the urge type, and childbearing does not appear to be an important risk factor. Overweight women appear to be at a higher risk of developing incontinence in midlife.


   ACKNOWLEDGMENTS
 
SWAN has grant support from the National Institutes of Health, Department of Health and Human Services, through the National Institute on Aging, the National Institute of Nursing Research, and the National Institutes of Health Office of Research on Women's Health (grants NR004061, AG012505, AG012535, AG012531, AG012539, AG012546, AG012553, AG012554, AG012495). The supplemental funding source for this secondary analysis from the University of California Davis Health System Research Award is gratefully acknowledged.

Clinical Centers: University of Michigan, Ann Arbor, Michigan—MaryFran Sowers, Principal Investigator; Massachusetts General Hospital, Boston, Massachusetts—Robert Neer, Principal Investigator 1995–1999; Joel Finkelstein, Principal Investigator 1999–present; Rush University, Rush University Medical Center, Chicago, Illinois—Lynda Powell, Principal Investigator; University of California, Davis/Kaiser, California—Ellen Gold, Principal Investigator; University of California, Los Angeles, California—Gail Greendale, Principal Investigator; University of Medicine and Dentistry–New Jersey Medical School, Newark, New Jersey—Gerson Weiss, Principal Investigator 1995–2004; Nanette Santoro, Principal Investigator 2004–present; and the University of Pittsburgh, Pittsburgh, Pennsylvania—Karen Matthews, Principal Investigator. NIH Program Office: National Institute on Aging, Bethesda, Maryland—Marcia Ory 1994–2001; Sherry Sherman 1994–present; National Institute of Nursing Research, Bethesda, Maryland—Program Officers. Central Laboratory: University of Michigan, Ann Arbor, Michigan—Daniel McConnell (Central Ligand Assay Satellite Services).

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Hannestad YS, Rortveit G, Sandvik H, et al. (2000) A community-based epidemiological survey of female urinary incontinence: the Norwegian EPINCONT study. Epidemiology of Incontinence in the County of Nord-Trondelag. J Clin Epidemiol 53:1150–7.[CrossRef][Web of Science][Medline]
  2. Brown JS, Seeley DG, Fong J, et al. (1996) Urinary incontinence in older women: who is at risk? Study of Osteoporotic Fractures Research Group. Obstet Gynecol 87:715–21.[CrossRef][Web of Science][Medline]
  3. Brown JS, Grady D, Ouslander JG, et al. (1999) Prevalence of urinary incontinence and associated risk factors in postmenopausal women. Heart & Estrogen/Progestin Replacement Study (HERS) Research Group. Obstet Gynecol 94:66–70.[CrossRef][Web of Science][Medline]
  4. Alling Møller L, Lose G, Jørgensen T. (2000) Risk factors for lower urinary tract symptoms in women 40 to 60 years of age. Obstet Gynecol 96:446–51.[CrossRef][Web of Science][Medline]
  5. Chiarelli P, Brown W, McElduff P. (1999) Leaking urine: prevalence and associated factors in Australian women. Neurourol Urodyn 18:567–77.[CrossRef][Web of Science][Medline]
  6. Sowers M, Crawford S, Sternfeld B, et al. (2000) SWAN: a multicenter, multiethnic, community-based cohort study of women and the menopausal transition. In Lobo RA, Kelsey JL, Marcus R (Eds.). Menopause: biology and pathobiology(Academic Press, San Diego, CA) pp. 175–88.
  7. McHorney CA, Ware JE Jr, Raczek AE. (1993) The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 31:247–63.[Web of Science][Medline]
  8. Radloff LS. (1977) The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1:385–401.[CrossRef]
  9. Baecke JA, Burema J, Fritjers JE. (1982) A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr 36:936–42.[Abstract/Free Full Text]
  10. Sherbourne CD and Stewart AL. (1991) The MOS Social Support Survey. Soc Sci Med 32:705–14.[CrossRef][Web of Science][Medline]
  11. Moos RH, Fenn CB, Billings AG, et al. (1988) Assessing life stressors and social resources: applications to alcoholic patients. J Subst Abuse 1:135–52.[Medline]
  12. Barsky AJ, Goodson JD, Lane RS, et al. (1988) The amplification of somatic symptoms. Psychosom Med 50:510–19.[Abstract/Free Full Text]
  13. Sampselle CM, Harlow SD, Skurnick J, et al. (2002) Urinary incontinence predictors and life impact in ethnically diverse perimenopausal women. Obstet Gynecol 100:1230–8.[CrossRef][Web of Science][Medline]
  14. Gold EB, Sternfeld B, Kelsey JL, et al. (2000) Relation of demographic and lifestyle factors to symptoms in a multiracial/ethnic population of women 40–55 years of age. Am J Epidemiol 152:463–73.[Abstract/Free Full Text]
  15. Jackson RA, Vittinghoff E, Kanaya AM, et al. (2004) Urinary incontinence in elderly women: findings from the Health, Aging, and Body Composition Study. Obstet Gynecol 104:301–7.[Web of Science][Medline]
  16. Diokno AC, Brock BM, Brown MB, et al. (1986) Prevalence of urinary incontinence and other urological symptoms in the noninstitutionalized elderly. J Urol 136:1022–5.[Web of Science][Medline]
  17. Fultz NH, Herzog AR, Raghunathan TE, et al. (1999) Prevalence and severity of urinary incontinence in older African American and Caucasian women. J Gerontol A Biol Sci Med Sci 54:M299–303.[Abstract]
  18. Bump RC. (1993) Racial comparisons and contrasts in urinary incontinence and pelvic organ prolapse. Obstet Gynecol 81:421–5.[Web of Science][Medline]
  19. Duong TH and Korn AP. (2001) A comparison of urinary incontinence among African American, Asian, Hispanic, and white women. Am J Obstet Gynecol 184:1083–6.[CrossRef][Web of Science][Medline]
  20. Day Baird D, Dunson DB, Hill MC, et al. (2003) High cumulative incidence of uterine leiomyoma in black and white women: ultrasound evidence. Am J Obstet Gynecol 188:100–7.[CrossRef][Web of Science][Medline]
  21. Kjerulff KH, Guzinski GM, Langenberg PW, et al. (1993) Hysterectomy and race. Obstet Gynecol 82:757–64.[Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 NEJMHome page
L. L. Subak, R. Wing, D. S. West, F. Franklin, E. Vittinghoff, J. M. Creasman, H. E. Richter, D. Myers, K. L. Burgio, A. A. Gorin, et al.
Weight Loss to Treat Urinary Incontinence in Overweight and Obese Women
N. Engl. J. Med., January 29, 2009; 360(5): 481 - 490.
[Abstract] [Full Text] [PDF]

Home page
 JAMAHome page
I. Nygaard, M. D. Barber, K. L. Burgio, K. Kenton, S. Meikle, J. Schaffer, C. Spino, W. E. Whitehead, J. Wu, D. J. Brody, et al.
Prevalence of Symptomatic Pelvic Floor Disorders in US Women
JAMA, September 17, 2008; 300(11): 1311 - 1316.
[Abstract] [Full Text] [PDF]

Home page
 Am J EpidemiolHome page
S. L. Tennstedt, C. L. Link, W. D. Steers, and J. B. McKinlay
THE AUTHORS REPLY
Am. J. Epidemiol., July 15, 2008; 168(2): 234 - 235.
[Full Text] [PDF]

Home page
 Diabetes CareHome page
J. M. Lawrence, E. S. Lukacz, I.-L. A. Liu, C. W. Nager, and K. M. Luber
Pelvic Floor Disorders, Diabetes, and Obesity in Women: Findings from the Kaiser Permanente Continence Associated Risk Epidemiology Study
Diabetes Care, October 1, 2007; 30(10): 2536 - 3541.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
165/3/309    most recent
kwk018v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Waetjen, L. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Waetjen, L. E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?