| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a Health Center of Cabra, Córdoba, Spain
b Department of Statistics, University of Córdoba, Spain
c Geriatrics Unit, University Hospital, Guadalajara, Spain
d Epidemiology and Biostatistics Department, National School of Health, Madrid, Spain
e Association of Health Districts of Southern Córdoba, Spain
F.J. Gavira Iglesias, Urbanizaci\|[oacute]\|n Los Poleares, 23, 14900 Lucena, C\|[oacute]\|rdoba, Spain E-mail: med015490{at}nacom.es.
Decision Editor: William B. Ershler, MD
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. A cross-sectional study was made by means of a home health interview in a representative sample of people aged 65 years and older in the Basic Health Zone of Cabra (Córdoba). A questionnaire was used with two questions to detect urinary incontinence and further questions regarding its duration, frequency, volume, severity and psychosocial impact. Data were analyzed with basic statistics. A logistic regression model was created to identify which factors affect people's lifestyles.
Results. Among the 827 respondents the prevalence of incontinence reached 36% [95% confidence interval (CI): 34%–38%] and was higher (p < .001) in women (42%; 95% CI: 39%–45%) than in men (29%; 95% CI: 25%–38%). The most frequent duration of incontinence was from 1 to 5 years (16.8%). Fifteen percent had daily episodes and 10.6% estimated the volume of urine loss as high. Incontinence was severe in 9% of the affected subjects. Half the incontinent subjects expressed social limitations and negative feelings related to incontinence, and 21% admitted that incontinence negatively affected their lives. In the logistic regression model, both volume of urine loss 
and being married 
were the factors statistically associated with a negative influence of incontinence on lifestyle.
Conclusions. Urinary incontinence is a very frequent symptom in our population. Half the incontinent subjects suffer from some degree of psychosocial limitation. The volume of urine loss and being married are the most important factors with psychosocial impact.
URINARY incontinence is a frequent health problem and a significant cause of disability and dependence in the older population. In community-residing subjects, prevalence rates ranging from 5% to 54% have been reported (1)(2)(3)(4)(5)(6)(7)(8). These figures show great variations that are explained partly by the lack of uniformity in the definition of urinary incontinence, the period of time considered when incontinence was recorded (now as opposed to within the previous 2 or 12 months), the type of population studied (excluding or not subjects in nursing homes), and the methodology used (postal questionnaires as opposed to personal interviews) (7)(8).
Important consequences of incontinence on physical and psychological health and social relations of the sufferers have been described (2)(4)(6)(9)(10).
In Spain one previous epidemiological study (11) has been carried out on the prevalence and characteristics of incontinence in the elderly residing in the community. The purpose of the present study was to investigate the prevalence, duration, severity, and psychosocial impact of urinary incontinence in an elderly rural population.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
65 years (16.5% of a total population of 5,139 subjects according to the Municipal Census of 1991). People institutionalized in nursing homes were not excluded.
To perfect the questionnaire and estimate sample size, we first conducted a pilot study on a nonproportional random sample of 5% of the 5,139 subjects 
, assigning a third of the sample to each of the three age strata (65–74, 75–84, and 85 years). This distribution was selected to include all or most of the subjects pertaining to the oldest groups. Once prevalence data by age group were known from the pilot study, we obtained the final size of each age group by interviewing the required number of subjects after considering those already included in the pilot study. The final sample size and stratification by age groups was determined to estimate the prevalence with a precision of ±3% and a confidence interval (CI) of 95%.
A letter was sent to each individual 1 week before the interview. If the selected person was hospitalized, the interview took place 30 days after discharge. Three interviewers specifically trained for the study (one male general physician and two female nurses), who were members of Basic Health Zone where the study was carried out, collected study data. A caregiver or relative was allowed to attend the interview. In the case of subjects suffering from deafness, aphasia, or cognitive dysfunction, the information from the caregiver was considered valid and the questions about psychosocial impact were not asked.
Two questions were used to detect urinary incontinence: "Do you ever have involuntary or unexpected leakages of urine without being able to control them?" and "Do you ever wet or dampen your underwear, clothes, or bedclothes against your will?" We defined incontinence as a positive response to either of the two questions and/or the use of an urethral catheter (8)(12) or absorbent pads. We did not attempt to identify the type of incontinence. Only incontinent people completed the rest of the questionnaire. The duration of urinary incontinence was measured in intervals (less than 1 month, 1–12 months, 1–5 years, and more than 5 years). Severity of incontinence was graded with the Sandvik Scale (12), which is created by multiplying the reported frequency 
by the volume 
of the urine leakages. The resulting index value (from 1 to 8) was further categorized into slight (1 or 2), moderate (3 to 4), and severe (6 to 8). Consequently, severe incontinence means larger amounts at least once a week.
We evaluated psychosocial impact by asking subjects how important the leakages were, how they felt about them, what effect the bladder problems had on their lifestyle, what social restrictions were caused by incontinence (6), and what was the worst effect on their life (13).
The questionnaire was complemented by sociodemographic and functional information. The latter was evaluated with the Barthel index (BI) modified by Shah and colleagues (14), from which we excluded the urinary continence item, so that the maximum score was 90 points. The subjects were classified under total independence 
, moderate dependence 
, severe dependence 
and total dependence (BI 
20).
The data were analyzed with the BMDP and EViews statistical programs. Basic statistics were computed (average and standard deviation for quantitative data and percentages for qualitative data). Two logistic regression models were elaborated: one to estimate incontinence prevalence according to age and sex and another to detect risk factors for an adverse impact of incontinence on the lifestyle of incontinent subjects. Because of the sampling format used, weighting factors were applied to reestablish proportionality in each age group according to their real size within the general population, when expressing overall results. Estimation of the proportion p of urinary incontinence must take into account the experimental design used: nonproportional stratified sampling in three age groups; let 
be the population size, 
the number of persons in each age strata, and 
the respective sample sizes. The overall estimator of p is
 |
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The most notable sociodemographic and functional characteristics of the sample are the large proportion of widowed women, the high percentage of illiterate or uneducated people, the low number of people institutionalized in nursing homes, and the high prevalence of moderate dependence in self-care (Table 1 ).
|
The prevalence of incontinence by age groups (Fig. 1) was greater in women than in men, with the exception of the 75- to 79-year-old and 90-year-old age groups, in which it is similar. There is an average prevalence increase of 5% per 5-year interval (approximately 1% per year of age). A significantly higher average prevalence was observed in women compared with that in men 
.
|
|
|
or older age groups (p < .01) were detected.
|
|
, daily frequency of leakages 
and duration of incontinence greater than 5 years 
were significantly associated with adverse effects on the lifestyles of incontinent subjects. Age, gender, and marital status were not associated with the adverse effects of incontinence. In the multivariate analysis, the relation between an adverse impact of incontinence and being married reached statistical significance and the relation with leakage volume was similar in magnitude to that found in the bivariate analysis. Independent people in self-care or those with slight disability showed a greater impact of incontinence on their lifestyle according to multivariate analysis 
.
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The previous Spanish study (11) was carried out in 1996 in both male and female subjects 65 years of age, with a similar sampling format and home interview. In contrast to our study, their survey was performed by non–health care interviewers and included only one question to detect incontinence. The population studied was urban and excluded institutionalized elderly. The response rate was 71%, and the overall prevalence reached 15.5%; no difference was seen between genders.
Few population surveys have determined the prevalence of incontinence in the elderly of both sexes (7)(15). Previous studies show variations in prevalence levels (1)(2)(3)(4)(6)(7)(15) that are in some cases explained by the different criteria used for diagnosis of incontinence (7)(15). Other studies, which introduce restrictions in the definition of incontinence (2)(6), show a clearly lower prevalence of urinary incontinence. The prevalence level in our population was 36% and is similar to that (30%) observed by Diokno and colleagues (3), who also used a broad definition of incontinence as well as two questions to identify urinary incontinence, a tool that probably allows slight or moderate incontinence to be detected in contrast to surveys that include only a single question (1)(4)(5)(6)(11). The slightly higher prevalence rates recorded in our study compared with the data of Diokno and colleagues could be due to several factors, such as the inclusion of older subjects, the higher response rate, and the presence of relatives in 31% of the interviews who reduced the probability of hiding the problem and allowed more dependent people to be included.
We found that prevalence was 1.4 times higher in women than in men. As in other studies (1)(3)(6), the difference in incontinence prevalence rates between women and men diminishes or disappears toward the end of life. In summary, our prevalence data in men and women are within the range found by Thom (15) and further evaluated by Hampel and colleagues through meta-analysis (7) in population-based studies carried out by personal interviews.
As in other studies of elderly populations (2)(4), the prevalence of incontinence increased with age. Women from 75 to 79 years of age are an exception in that the prevalence decreased significantly, a finding that has been previously pointed out (2) and might reflect the natural progression of stress-related incontinence, which diminishes in women of advanced age (2)(15)(16)(17).
In contrast with other reviewed studies (15), in our study we found a high incidence (15.4%) of daily incontinence, especially in men (13.7%), in whom the condition might be related to prostatic disorders.
Despite the lack of agreement on the definition of severe incontinence, its prevalence is consistent among published studies (7)(8)(18). Severe incontinence was seen in 9% of our subjects, a prevalence that falls within the range considered reasonable for elderly subjects (5%–12% of subjects) (7)(8)(18).
The International Continence Society definition of urinary incontinence (involuntary loss of urine, which is objectively demonstrable and causes a social or hygienic problem) (19) adds dimensions of severity and consequences to the phenomenon of involuntary loss of urine. The emotional distress, social restrictions, and practical inconveniences caused by incontinence are thus part of its definition. However, when the International Continence Society's definition is used, the prevalence of incontinence may be substantially reduced. In our study, only 10% of the evaluated subjects considered that incontinence significantly affected their lives and only a quarter perceived the problem as being quite or very important; 43% reported psychological distress, over half reported restrictions in relation to incontinence, and 65% admitted some practical inconvenience.
Brocklehurst (6) carried out a study in the UK comparable with ours from a methodological point of view (cross-sectional survey, men and women over 30 years of age, personal interview). The psychosocial impact of incontinence seems to affect a lower proportion of our subjects compared with those of Brocklehurst in terms of incontinence perceived as important/very important (23% vs 35%), important effect on lifestyle (10% vs 36%), and psychological distress (43% vs 60%), but social restrictions were higher in our population (52% vs 45%).
On the other hand, our subjects were more affected in relation to concrete aspects of the psychosocial impact (feelings, social restrictions, and practical inconveniences) than to its generic aspects (the importance the subject gives to incontinence and the general effect on his or her lifestyle). Nevertheless, the real importance of the problem could be underestimated in our study because of some degree of occultation by the patient of the consequences of incontinence and the facts that psychosocial impact was not measured in 34 patients and that study methodology allowed for the detection of a great number of slight incontinence sufferers. In agreement with Mitteness (20) we suspect that the low cultural level and resigned or fatalistic attitudes of our population ("the belief that urine losses are the result of aging and nothing can be done about them") may explain the apparently low concern of our subjects about incontinence. However, these observations prompt us to underline the complexity of evaluating the different spheres considered within the quality-of-life concept in incontinent subjects.
According to the multivariable analysis, the volume (not frequency) of urine losses and being married were variables significantly associated with the impact of incontinence on the quality of life of the affected subjects (21)(22)(23).
Some limitations of the study should be taken into consideration when our results are interpreted. First, objective checks of the presence of incontinence or measurements of its frequency and volume were not performed. Nevertheless, there is evidence that the prevalence of incontinence defined by objective measurements differs very little from that found by means of a questionnaire (9)(18). Second, we did not estimate the interviewer to interviewer consistency on the subject's responses. However, our interviewers did go through a common training process on the evaluation system to be followed during the interview.
Primary care professionals should include specific questions about incontinence when assessing older patients and, once diagnosed, determine the severity and type of incontinence, and finally ensure, either personally or through specialized units, that appropriate care and follow-up are given to the patient.
|
Acknowledgments |
|---|
Received March 8, 1999
Accepted August 29, 1999
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. E. Morley Editorial: Drugs, Aging, and the Future J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2002; 57(1): M2 - 6. [Full Text] [PDF]
|
|
|
|
|
|
S. Maggi, N. Minicuci, J. Langlois, M. Pavan, G. Enzi, and G. Crepaldi Prevalence Rate of Urinary Incontinence in Community-Dwelling Elderly Individuals: The Veneto Study J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2001; 56(1): 14M - 18. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
+e,
for subjects with incontinence and 
for those without incontinence. The z statistics for the model coefficients were -6.21, 4.81, and 3.36, all of them with p < .001. The Akaike information criteria was 1.306.