Abstract
Objective. To describe the prevalence of urinary incontinence and to find characteristics useful in general practice for identifying middle-aged and elderly women with the problem. Design. Cross-sectional interview study. Setting. Population-based samples of Danes. Subjects. A total of 5795 women older than 45 years (46+ years). Main outcome measures. Prevalence of incontinence and clinical characteristics assessed by standardized interview questions. Results. The overall prevalence of urinary incontinence was 20% among women less than 60 years of age and 44% among those older than 80 years. Increasing age was highly associated with both forms of incontinence (stress and urge). High body mass index (BMI), chronic lung disease, and stroke were also associated with both forms, while number of children was associated with stress incontinence only. Predictive models show that 56% of women characterized by high age (older than 80 years) and overweight (BMI higher than 30) will suffer from urinary incontinence. The low-risk group defined by these two parameters (aged 46–60 years and not overweight) still had a 19% prevalence in the last month. Conclusion. The prevalence of urinary incontinence increased with age. Even in the low-risk groups the problem was very common in old age. Questions about incontinence should, therefore, be asked in relevant consultations with all elderly female patients.
The International Continence Society has defined urinary incontinence as the complaint of any involuntary leakage of urine Citation[1]. The two major forms are stress incontinence (defined as the complaint of involuntary leakage on effort or exertion, or on sneezing or coughing) and urge incontinence (defined as the complaint of involuntary leakage accompanied or immediately preceded by urgency). Urinary incontinence is a common condition, with an estimated prevalence in the range of 15–30% among middle-aged women Citation[2–4]. The problem is apparently even more frequent among the elderly Citation[2–7], but the epidemiology and risk factors of incontinence are less well described for women above 60 years of age.
Incontinence may be socially stigmatizing Citation[8], and women are often reluctant to talk about this problem. Many are therefore not treated although minimal care programmes have been developed and provide a good prognosis. Up to 70% of patients seeking help may be cured or improved by simple methods in general practice Citation[9]. General practitioners (GPs), therefore, play a major role in the diagnosing of women suffering from urinary incontinence and in offering guidance and treatment. In a situation where only a minority of women with incontinence problems consults doctors Citation[8], Citation[10–12], GPs may need to take the initiative. It may, therefore, be of value to identify clinical characteristics that are easily observed and highly predictive of incontinence.
Based on interviews with a large population-based sample of middle-aged and old Danish women the present study aimed at describing the prevalence of urge and stress incontinence. Furthermore, we wanted to find simple clinical characteristics that are useful in general practice for identifying women with a high probability of urinary incontinence.
Material and methods
Study population
The study population comprised 5795 women sampled in three surveys which used the same questionnaire instrument: (1) 2075 women from a random sample of 46- to 68-year-old Danish twins (Longitudinal study of Middle-Aged Twins 1999) Citation[13], Citation[14], (2) 2400 women from a sample of all 70- to 94-year-old Danish twins (Longitudinal Study of Aging Danish Twins – 1995 (n=1399), 1997 (n=354), 1999 (n=647)) Citation[15], Citation[16], and (3) 1320 women from a sample of all Danes born in 1905 and living in 1998 (The 1905 cohort of Danes) Citation[17]. The participation in the three cohorts ranged from 63% to 83% Citation[16]. A total of 574 women (41 women from the middle-aged twin cohort, 185 women from the old twin cohort, and 358 women from the 1905 cohort) who were unable to complete the interview without proxy were excluded from the analysis. The studies were approved by the Danish Data Protection Agency and the Medical Ethics Committee for Funen and Vejle Counties, and conducted in accordance with the Helsinki Declaration and Danish law.
Interviews
The residence of all the informants in the study population was identified through the Central Person Registry. The women received a letter explaining the study and asking permission for an interviewer to come to their home in order to conduct a health-related one-hour face-to-face interview.
Interviewers from the Danish National Institute of Social Research were used for all the study populations. The interviewers were not medically trained but had substantial experience in interviewing the elderly. All interviewers participated in a detailed training programme and were closely monitored during the periods of interviewing. Only when at least three unsuccessful attempts to contact an informant at her residence at different times had been made was the informant considered to be impossible to contact.
Questions
Female urinary incontinence was identified by means of two questions, which had shown good reproducibility and external validity against a gold standard (open interview about symptoms from the urinary tract) Citation[18]. Before asking the two questions the interviewers gave a short introduction: “Many women have problems with involuntary loss of urine. Some become incontinent in connection with physical activity and some because of an irresistible urge to void. I will ask you a couple of questions about this. Please think of your experience within the past month”. They then asked a question about stress incontinence: “During the past month have you involuntary been wetting yourself in connection with physical exertion, e.g. coughing, lifting, sneezing or laughing?” (answers: “No”, “Yes, once during the past month”, “Yes, several times but less than once a week”, “ Yes, once or several times a week, but not daily”, “Yes, daily”), and one about urge incontinence: “During the past month, have you experienced such a strong urge to pass water that you had trouble getting to the toilet in time?” ( “No”, “Yes, once during the past month”, “Yes, several times but less than once a week”, “ Yes, once or several times a week, but nor daily”, “Yes, daily”). Women who reported more than one leakage during a month were classified as suffering from urinary incontinence, either stress and/or urge.
The interviews also produced data that could be used to characterize patients with incontinence. Characteristics of patients were sought among factors that traditionally have been associated with urinary incontinence Citation[19–23], i.e. age, body mass index (BMI), number of children, use of diuretics, hormone replacement therapy, previous myocardial infarction, stroke or hysterectomy, and presence of a number of chronic conditions (diabetes, Parkinsonism, and lung disease). Only characteristics likely to be known by the GP were included in the analysis while “hidden” risk factors such as a family history of incontinence were excluded. Classification of the medical conditions was based on the patient's information only. Age was grouped as “ < 60 years”, “60–80 years” and “ >80 years”. BMI was presented as “ < 25”, “25–30” and “ > 30”. Number of children was presented as “No children”, “1–2”, and “3 + ”. All other characteristics were dichotomized as “present” or “not present”.
Analysis
The association between urinary incontinence and patient characteristics was analysed using logistic regression with robust variance analysis to adjust for twin similarity Citation[24]. All estimates are presented with 95% confidence intervals. Only those clinical characteristics that appeared to have a relevant bivariate association with incontinence were included in the model. We made marginal prediction of prevalence illustrating how these characteristics can be used to classify women into low- and high-risk groups. The marginal prediction is based on estimated coefficients and specific values of covariates. Those covariates that are not set to a specific value in a given prediction are all set to their mean value as observed in the data. The statistical software Stata 8.0 (Stata Corporation) was used for all analyses.
Results
Almost every third of the women in this study were suffering from urinary incontinence with the prevalence increasing with age from 20% below age 60 years to 44% among women aged 80+ years (). The median age of the population examined was 75 years (61 and 88 years as inter-quartile range [IQR]), median number of children was 2 (1; 2 IQR), and the median of BMI was 23 (21; 26 IQR). Most of the analysed factors, but not “hormone replacement therapy” and “hysterectomy”, had a strong association with incontinence in the bivariate analysis as illustrated by the confidence intervals in .
Table I. “Urinary incontinence” (stress or/and urge), “stress incontinence”, and “urge incontinence” in relation to a number of clinical characteristics.
gives odds ratios for those factors associated with urinary incontinence. Increasing age was a characteristic associated with both forms of incontinence and especially urge incontinence (odds ratio 4.0). A high BMI, chronic lung disease, and stroke were characteristics with almost equal odds ratios for stress and urge incontinence (odds ratios around 1.7). Diuretics, diabetes, and Parkinsonism were to some degree associated with both forms of urinary incontinence. The odds ratios did not change significantly when hormone replacement therapy and hysterectomy were included in the analysis. The number of children was a characteristic for stress incontinence only. and combined show that most of the factors associated with urinary incontinence are common.
Table II. Odds ratios for the association between clinical characteristics and “urinary incontinence” (stress or/and urge), “stress incontinence” and “urge incontinence”.
A predictive model showed that 56% of women characterized by high age (older than 80 years) and overweight (BMI higher than 30) (n=112) will suffer from some form of urinary incontinence while 19% of women in a corresponding low-risk group (aged 46–60 years and no overweight) (n=851) will suffer from urinary incontinence. Adding common chronic diseases to this model increased the probability of incontinence. As an example, chronic lung disease increased the likelihood of urinary incontinence by 10% in the high-risk group. Also, for the specific forms of urinary incontinence the clinical characteristics were highly discriminatory as the prevalence of urge was more than five times higher among the women in this high-risk group compared with women with low risk. For stress incontinence the prevalence increased somewhat less (by a factor of 3).
Discussion
Urinary incontinence is very frequent among elderly women. General practitioners should be aware that women characterized by high age and high BMI will be very likely to suffer from urinary incontinence. Also chronic lung disease, diabetes, diuretic consumption, and stroke were characteristics associated with incontinence.
Our study was based on a large population of elderly women sampled in three different studies all with relatively high participation rates (63–83%). Several reminders were used in this group of women of high age. Dropout is, however, likely to be more frequent among the frail elderly, who are also more likely to have incontinence problems. Our estimates of prevalence may therefore to some degree be underestimated, while associations between patients’ characteristics and incontinence are less likely to be biased by dropout. The three patient populations were nearly mutually exclusive with respect to age and calendar year of inclusion, and hence we cannot exclude calendar time effects. Such an effect is, however, likely to be minor due to the relatively short sampling period compared with the typically long duration of incontinence problems. Some of the participants in the study were sampled as twins, but this is not likely to make them less representative of the larger population with regard to the questions analysed Citation[25].
The measure of incontinence used in the present study relies on the individual's subjective assessment of whether she is incontinent or not. This is in accordance with the new definition of the International Continence Society Citation[1]. According to previous definitions the condition should be demonstrated objectively and be of psychological importance to the patient but these criteria have been omitted in the new definition, which made it much more operational for epidemiological studies. We used two standardized questions to identify women with urinary incontinence. The questions have been validated against a long open interview and compared with a prospective registration of leakages Citation[18]. Recall bias must, however, be taken into consideration. Some of the participants were old and parts of the questionnaire related to occurrences way back in the women's lives, e.g. deliveries. Also the women's information about medical conditions may include some misclassification, which may reduce correlations between the various conditions and incontinence. Exclusion of women answering by proxy in the interview was employed to reduce recall bias and biases in general. BMI and age were estimated from reliable data obtained in connection with the interviews.
The estimates of the prevalence of urinary incontinence found in this study are comparable with results from several other studies Citation[2–7], taking into account that differences in the definitions of urinary incontinence may produce some variation Citation[1], Citation[2]. The purpose of the present study was not to look for the aetiology of urinary incontinence but to describe characteristics that would make it easy for GPs to identify women with a high probability of incontinence. We confirmed the association between urinary incontinence and age, BMI, stroke, chronic lung disease, diabetes, and diuretic consumption demonstrated in previous studies Citation[19–23]. In contrast to most other studies we only found a rather weak association between parity and stress incontinence. Our finding is, however, in accordance with another recent study of women above the age of 65 years, which found little effect of parity among the elderly Citation[26].
The focus of the clinical literature regarding urinary incontinence has been on young women. Incontinence is, however, far more common among the elderly. Our predictive model showed that almost 20% of women in the lowest risk group (aged 46–60 years and not overweight) had experienced incontinence within the last month. A major number of women with urinary incontinence will therefore be overlooked, if GPs restrict case-finding to the smaller groups that have a very high prevalence of incontinence (e.g. the oldest old with high BMI and chronic diseases). In relevant consultations all elderly women should be asked about incontinence. GPs will have varying definitions of such consultations. It should, however, be stressed that some of the problems associated with case-finding in general practice, e.g. stigmatizing a patient with an unknown disease, do not apply to incontinence. The problem is already known to the patient. It is only the GP who is unaware, and urinary incontinence may often be improved by minimal care programmes.
Key Points
Urinary incontinence is a treatable condition, which often goes undiagnosed.
A total of 56% of 80+ year-old women with a high body mass index (>30) had urinary incontinence, and the prevalence increased further if chronic disease was present.
Even in our low-risk group (aged 46–60 years and not overweight) 19% had experienced incontinence within the last month.
The problem should, therefore, be raised in relevant consultations with all elderly female patients.
Related Research Data
References
- Abrams P, Cardozo L, Fall M, Griffiths D, Rosier P, Ulmsten U, et al. The standardisation of terminology of lower urinary tract function. Report from the Standardisation Sub–committee of the International Continence Society. Neurourol Urodyn 2002; 21: 167–78
- Hunskaar S, Burgio K, Diokno A, Herzog AR, Hjälmås K, Lapitan MC. Epidemiology and natural history of urinary incontinence in women. Urology 2003; 62: 16–23
- Hannestad YS, Rortveit G, Sandvik H, Hunskaar S. A community-based epidemiological survey of female urinary incontinence: The Norwegian EPICONT Study. J Clin Epidemiol 2000; 53: 1150–7
- Stoddart H, Donovan J, Whitley E, Sharp D, Harvey I. Urinary incontinence in older people in the community: a neglected problem?. Br J Gen Pract 2001; 51: 548–54
- Perry S, Shaw C, Assassa P, Dallosso H, Williams K, Brittain KR, et al. An epidemiological study to establish the prevalence of urinary symptoms and felt need in the community: The Leicestershire MRC Incontinence Study. J Public Health Med 2000; 23: 427–34
- Bjornsdottir LT, Geirsson RT, Jonsson PV. Urinary incontinence and urinary tract infections in octogenarian women. Acta Obstet Gynecol Scand 1998; 77: 105–9
- Nuotio M, Jylha M, Luukkaala T, Tammela TL. Urinary incontinence in a Finnish population aged 70 and over. Scand J Prim Health Care 2003; 21: 182–7
- Largro-Janssen TLM, Smits AJA, Van Weel C. Women with urinary incontinence: Self-perceived worries and general practitioners’ knowledge of the problem. Br J Gen Pract 1990; 40: 331–4
- Seim A, Sivertsen B, Eriksen BC, Hunskaar S. Treatment of urinary incontinence in women in general practice: Observational study. BMJ 1996; 312: 1459–62
- 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
- Teunissen D, Largo-Janssen T. Urinary incontinence in community dwelling elderly: Are there sex differences in help-seeking behaviour?. Scand J Prim Health Care 2004; 22: 209–16
- Hägglund D, Walker-Engström M-L, Larsson G, Leppert J. Changes in urinary incontinence and quality of life after four years. Scand J Prim Health Care 2004; 22: 112–7
- Kyvik KO, Christensen K, Skytthe A, Harvald B, Holm NV. The Danish Twin Register. Dan Med Bull 1996; 43: 467–70
- Gaist D, Bathum L, Skytthe A, Jensen TK, McGue M, Vaupel JW. Strength and anthropometric measures in identical and fraternal twins: no evidence of masculinization of females with male co-twins. Epidemiology 2000; 11: 340–3
- Christensen K, McGue M, Yashin A, Iachine I, Holm NV, Vaupel JW. Genetic and environmental influences on functional abilities in Danish twins aged 75 years and older. J Gerontology 2000; 55: 446–52
- Christiansen L, Frederiksen H, Schousboe K, Skytthe A, Von-Vurmb-Schwark N, Christensen K, et al. Age- and sex-differences in the validity of questionnaire-based zygosity in twins. Twin-Res 2003; 6: 275–8
- Nybo H, Gaist D, Jeune B, Bathum L, McGue M, Vaupel JW, et al. The Danish cohort: A genetic-epidemiological nationwide survey. J Aging Health 2001; 13: 32–46
- Rohr G, Christensen K, Ulstrup K, Kragstrup J. Reproducibility and validity of simple questions to identify urinary incontinence in elderly women. Acta Obstet Gynecol Scand 2004; 83: 1–4
- Rortveit G, Dalveit AK, Hannestad YS, Hunskaar S. Urinary incontinence after vaginal delivery or cesarean section. N Engl J Med 2003; 348: 900–7
- Mommsen S, Foldspang A. Body mass index and adult female urinary incontinence. World J Urol 1994; 12: 319–22
- Brown JS, Sawaya G, Thom DH, Grady D. Hysterectomy and urinary incontinence: A systematic review. Lancet 2000; 356: 535–9
- Diokno A, Brock BM, Herzog AR, Bromberg J. Medical correlates of urinary incontinence in the elderly. Urology 1990; 26: 129–38
- Maggi S, Minicuci N, Langlois J, Pavan M, Enzi G, Crepaldi G. Prevalence rate of urinary incontinence in community-dwelling elderly individuals: The Veneto study. J Gerontol A Biol Sci Med Sci 2001; 56: 14–18
- Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika 1986; 73: 13–22
- Christensen K, Vaupel JW, Holm NV, Yashin AI. Mortality among twins after age 6: Fetal origins hypothesis versus twin method. BMJ 1995; 310: 432–6
- Rortveit G, Hannestad YS, Dalveit Ak, Hunskaar S. Age- and type-dependent effects of parity on urinary incontinence: The Norwegian EPICONT Study. Obstet Gynecol 2001; 98: 1004–10