Prevalence and Risk Factors of Urinary Incontinence in Frail Elderly Females

Background/Purpose. Urinary incontinence (UI) is an important geriatric health problem, and it is linked to frailty syndrome. We had conducted a study to detect the prevalence and risk factors of UI and its effect on quality of life (QOL) among frail elderly females living in Cairo, Egypt. Methods. We carried out a cross-sectional study on 130 frail elderly females sixty years and older, attending Ain Shams Geriatrics Hospital, Cairo, Egypt. Each patient gave oral consent and then was subjected to history taking, full clinical examination, diagnosis of frailty (clinical frailty scale), assessment of UI by the Arabic version of International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form (ICIQ-UI SF), assessment of QOL by using the Arabic version of Incontinence Impact Questionnaire Short Form (IIQ-7 SF), and complete urine analysis. Results. The prevalence of UI among the studied population was 80%. Mixed UI was the most prevalent type. UI was significantly associated with older age, functional impairment, multiparity, osteoarthritis, stroke, vaginal prolapse, and laxative use. All IIQ-7 subscales were higher (worse health-related QOL) for women with mixed UI. Conclusion. Urinary incontinence is prevalent in frail elderly females. Mixed UI, compared with other types, has a significant negative impact on all domains of quality of life.


Introduction
Urinary incontinence (UI) is defined by the International Continence Society as any involuntary leakage of urine. It is a common clinical problem, and its incidence increases with age [1]. UI is a common symptom in older people, and some reports suggest that UI is associated with frailty, whereby the ability of the body to cope with stress and physiological functions decreases [2].
Normal ageing is not a cause of UI, although age-related changes in lower urinary tract function can predispose older people to UI which is then exacerbated by comorbidities. UI is a major cause of disability and dependency, significantly increasing the risk of care home placement. It also predisposes to career negativity and stress, which itself is a major factor in placement for institutional care [1]. e elderly rank UI among the 4 most distressing disorders after angina, difficulty with ambulation, and psychiatric disorders [3]. UI is frequently associated with a negative impact on quality of life (QOL) of the patient, despite not being a life-threatening condition. UI has many physical and psychological effects on the patients, while at the same time, it is associated with an additional financial burden [4].
Evaluation of the effects of urinary symptoms on the quality of life (QOL) is an important issue. UI may lead to shame, loss of self-esteem, and social isolation [5].
Since incontinence is associated with an increased risk of a global functional impairment, in persons who become incontinent after the age of 65 years [6], this parameter may be an important early marker for signaling the onset of frailty, and the 4 th International Consultation on Incontinence has urged researchers to better understand the correlation between UI and frailty [7].
Evaluating the risk factors is an important component in the assessment of UI in the older adults. In both men and women, increased age, genetics, obesity, and tobacco use are associated factors [8]. In women, multiparity is also associated with UI [9].
Screening has the potential to identify urinary incontinence in many women who silently experience its adverse effects but may benefit from appropriate evaluation and treatment. Effective screening may lead to earlier treatment, including behavioral, medical, and surgical interventions, depending on the patient's age and the type and severity of symptoms [10]. us a cross-sectional study was conducted to find the prevalence and risk factors of UI and its effect on QOL among frail elderly females living in Cairo, Egypt.

Study Design and Participants.
is was a cross-sectional study. e study was carried out at Ain Shams Geriatrics Hospital in Cairo, Egypt. e study was approved by the ethical committee of Ain Shams University. e study sample comprised 130 frail females aged sixty years and above.
Exclusion criteria of patients are as follows: (

Geriatric Domains Assessment.
All participants were screened for dementia using the Arabic version by El-Okl et al., [11] of minimental state examination (MMSE) [12] and depression using the Arabic version by Shehata et al. [13] of geriatric depression scale (GDS) [14]. Functional assessment was done using activities of daily living (ADL) [15] and instrumental activities of daily living (IADL) scales [16].
Frailty was assessed by the clinical frailty scale (CFS). e clinical frailty scale is a reliable tool to identify frailty in clinical settings. e CFS ranges from 1 (very fit) to 9 (terminally ill) based on descriptors and pictographs of activity and functional status. According to CFS, study participants were divided into three categories: mildly frail (CFS 5), moderately frail (CFS 6), and severely frail (CFS 7) [17]. Nonfrail subjects (CFS 1-4) were excluded from our study. Of note, we did not include subjects approaching the end of life (CSF 8-9).
Assessment of QOL was done using the Arabic version of Incontinence Impact Questionnaire, short form (IIQ-7 SF) [19]. e IIQ-7 SF questionnaire is a seven-item questionnaire designed to assess different domains of QOL impairment. It has a four-point rating scale: 0 � never, 1 � mild, 2 � moderate, and 3 � severe; the higher the score, the poorer the QOL.
Laboratory testing included complete urine analysis. Pyuria was defined as the presence of at least 10 white blood cells per cubic millimeter of centrifuged urine by high power field [20].

Statistical Analysis.
e collected data were coded, tabulated, and statistically analyzed using IBM SPSS statistics (Statistical Package for Social Sciences) software version 22.0, IBM Corp., Chicago, USA, 2013.
Description of all data was done in the form of mean (M) and standard deviation (SD) for all quantitative variables. Frequency and percentage were used for all qualitative variables. Comparison between quantitative variables was done using t-test. Comparison of qualitative variables was done using the chi-square test. Correlations (r-value) were assessed by Spearman rank correlation to find relation between different variables. While positive r-value indicates direct correlation, negative r-value indicates inverse relationship between the variables. e one-way ANOVA test was used to compare between means of more than 2 study groups using the F value. P value < 0.05 was considered as statistically significant.

Results
e subjects surveyed in our study were 300 elderly females aged sixty years and above who attended Ain Shams Geriatrics Hospital, Cairo, Egypt, during the period from June 2018 till April 2019. Among them, we included 130 frail elderly females in our study which were selected according to the clinical frailty scale (CFS). e mean age of the frail elderly females was 70.7 ± 8.3 years. Among the 130 frail participants, we found 104 (80%) patients suffering from UI. So we actually included 130 frail participants (who met the frailty criteria), but calculations specific for incontinent subjects (e.g., relationship between type of UI and quality of life) included only 104 patients (i.e., those having any type of UI). is is given in Tables 1, 2, and 3, in which data of incontinent patients were analyzed. In contrast, Tables 4, 5, and 6 (comparing incontinent and continent subjects) included the whole 130 study participants. e prevalence of UI among the studied population was 80%; with a mean duration of 5.9 ± 2.6 years. Mixed UI was the most prevalent type (91%) among participants with UI.
ere was a significant relationship between the ICIQ-UI score and the type of UI, being most severe in those with mixed UI (Table 1).
ere was a statistically significant   relationship between ICIQ-UI and the duration of UI as there was a statistically significant higher mean duration of UI among patients with severe and very severe UI in comparison to moderate one according to the ICIQ-UI severity criteria (Tables 2).
As regards to quality of life (QOL) of affected participants using IIQ-7, there was a significant affection of all quality of life domains in patients with mixed UI, compared with other types of UI (Table 3). ere was a significant relationship between UI and the presence of depression (positive screening by GDS) ( Table 4).
ere was a highly significant relationship between UI and frailty (as detected by clinical frailty scale); all moderately and severely frail participants had UI (Table 5). ere was a highly significant relationship between frailty category and ICIQ-UI score, being more severe in those participants with more advanced frailty ( Table 6).
As regards demographic characteristics and risk factors, participants with UI, compared to those without, were significantly older, more commonly multiparas and had higher degree of functional impairment in both ADL and IADL (Table 7 and 8). As regards associated comorbidities, osteoarthritis, stroke, and vaginal prolapse were significantly more prevalent in participants with UI, compared to those without UI. As regards medication use, laxative use was significantly more prevalent in participants with UI, compared to those without UI.

Discussion
e study revealed that the prevalence of UI among the studied population was 80%. We found a highly significant relationship between UI and frailty. All moderately/severely frail participants in our study had some type of UI (P value < 0.001). In the same way, this study revealed a highly significant relationship between frailty category and severity of UI (ICIQ-UI score), being more severe in those participants with more advanced frailty (P value > 0.001).    In agreement with these results, a cross-sectional study done by Wang et al. [21] showed that frailty was more common among subjects with UI than those without UI (60.7% vs. 32.3%, P value < 0.001). Chong et al. [2] showed that among 210 participants (mean age 89.4 ± 4.6 years, 69.5% female, 50.0% frail), UI was present in 47.6%, with a higher prevalence among frail individuals (64.8% vs. 30.5%, P < 0.001). Incident UI was more common in frail participants (at discharge: 24.3% vs. 9.6%, P � 0.038; 6 months: 43.2% vs. 21.7%, P � 0.020; and 12 months: 56.8% vs. 33.3%, P � 0.020).
In the study of Kang and Kim [22], participants (n � 404) who visited a geriatric clinic were divided into two groups according to the presence or absence of UI (based on questionnaire results) and the relationship between each factor associated with physical frailty and the risk of UI was analyzed. e participants in the UI group showed a weaker grip strength and slower walking speed (P < 0.01 and P � 0.01, respectively) and had more experiences of unintentional weight loss and falls (P � 0.04 and P < 0.01, respectively).
Previous studies reported different prevalence rates of UI among elderly people. is heterogeneity can be explained by differences in participants' characteristics (e.g., sex, mean age, frailty status, and comorbidities), study setting (community, care homes, etc.), and methodological differences.
For example, it was reported that 15-30% of community-dwelling older people have UI. Many studies reported a higher prevalence of UI among care home residents, which ranged between 50% and 80% because UI is associated with older age, frailty, cognitive impairment, and limited mobility leading to a greater level of dependency [1].
As regards type of UI, mixed UI was the most prevalent type of UI among the studied population (91.35% of participants with UI). is large percentage was expected in this frail cohort suffering from functional impairment and multiple comorbidities, making them more likely to have more than one type of UI.
is agree with the study of Talley et al. [23] that showed the most common type of UI in frail women was mixed stress and urgency (62%), followed by urgency (22%), stress (14%), and functional (2%).
Wang et al. [21] conducted a cross-sectional study on 440 participants and found that prevalence of UI was 19.1%. 51.2% of the incontinent subjects had urge incontinence, and 41.7% had functional incontinence. e current study showed a significant relationship between type of UI and severity of UI (as determined by the ICIQ-UI score), being most severe in those with mixed UI. In agreement with current results, Schreiber Pedersen et al. [24] conducted a postal survey in two regions in Germany and Denmark, including 8000 adult women, and found that women with mixed UI reported a higher mean total ICIQ score in all age groups, than did women with urge UI or stress UI.
ere was also a significant relationship between the duration of UI and ICIQ-UI score, being most severe in those with longer duration of UI. e current study reveals that Incontinence Impact Questionnaire (IIQ-7) scores were higher for women with mixed UI. A significant affection of all quality of life domains was found in participants with mixed urinary incontinence, compared with those suffering from other types of UI. is agrees with the study of Frick et al. [25] who found that IIQ scores were higher for women with mixed incontinence versus urge or stress incontinence (P value < 0.01). Yu et al. [26] conducted a cross-sectional study to investigate the impact of UI on QOL using IIQ-7. eir study involved 1608 adult Taiwanese females. ey found that women with mixed UI had a higher IIQ-7 score compared to those with stress or urge UI.
is study revealed a significant affection of praying in patients with mixed UI, compared with other types of UI. Yet this result could not be compared to studies done in different non-Muslim populations because praying is actually not an item in the original IIQ and was added in the Arabic version to suit our population. e affection is expected to be influenced most by the amount and the frequency of urine loss. Studies in Muslim women revealed that UI led to the limitation of religious life, related to the need of cleanliness for the practice of prayer [27]. e current study also showed a significant affection of physical activities in patients with mixed UI, compared with other types of UI. In a similar way, Manonai et al. [28] found that mixed UI has the greater impact on physical activities (20.4%) than urge UI (11.5%) and stress UI (9.1%) in their cross-sectional study that involved 319 females with UI. ey measured the impact of UI on QOL using IIQ-7.
As regards affection of social life, we found a significant affection of social activities in patients with mixed UI, compared with other types of UI. Manonai et al. [28] reported similar results in their cross-sectional study. ey found that mixed UI has the greatest impact on social life (15.6%) followed by urge UI (10.3%) and stress UI (5.5%), respectively.
Moreover, the current study revealed marked affection of travelling in patients with mixed UI, compared with other types of UI. In agreement with this result, Manonai et al. [28] found that mixed UI has the greatest impact on travelling (25.3%) followed by urge UI (19.2%), with the lowest affection found in stress UI (7.3%).
e current study showed a significant relationship between UI and the presence of depressive symptoms (positive screening by geriatric depression scale, GDS) where 89% of depressed participants had UI (P value � 0.026). No significant relationship was found between the type of UI and the presence of depressive symptoms. One of the comorbidities associated with UI include depression [29].
As regards emotional wellbeing, significant affection (in the form of depression/hopelessness and anxiety/frustration) was found in patients with mixed UI, compared with other types of UI. is agrees with the results of the study of Manonai et al. [28]. e authors found that mixed UI has the greatest impact on emotional life (30.6%) followed by urge UI (23.1%) and the least affection was found in stress UI (16.4%). is also agrees with the cross-sectional study done by Wang et al. [21], in which the authors concluded that subjects with UI had more depressive symptoms (GDS-5, P value � 0.02).
Many cross-sectional studies reported the association between UI and depression. ese include a large US population-based cross-sectional study [29] and another smaller Japanese study [30].
In contrast, negative findings were reported in a Korean study by Song and Bae [31]. Another study by Ko et al. [32] did not found an association between UI and self-reported sadness. ese negative findings may reflect differences in definition of depression and patient selection criteria.
UI has many risk factors, which we inquired about in our study. For instance, there was a significant relationship between parity and the presence of UI. Most of the study participants with UI were multiparas (78.85%, P value � 0.005). An epidemiological study in Italy showed that vaginal birth increased the risk of stress UI (OR 3.8 vs. nulliparous) but not of urge UI [33]. Fabio et al. [34] confirmed these findings with regard to reproduction; history of vaginal births increased the risk of stress and mixed UI but not of urge UI and overactive bladder.
As regards obesity, the current study failed to prove the relation between UI and obesity. In contrast, Fabio et al. [34] found that the risk of all types of UI is increased in women with high body mass index.
is contradiction may be explained by the fact that our study included less number of obese subjects (38 participants, 29%). Of note, 28 of them had UI, while 10 only did not suffer from UI, but this difference did not reach statistical significance. e relationship between smoking and UI is unclear. Heavy current/former smokers have been shown to have a higher risk of UI [35]. Our data and that of others (Fabio et al. [34]; Burgio et al. [36]), however, have failed to confirm this relationship.
It is worthwhile mentioning that current study included a few number of smokers (16.92% of the whole study sample) because our population consisted of female participants only. Among 22 smokers included in our study, 19 had UI.
As regards functional limitation, participants with UI, compared to those without, had higher degree of functional impairment in both activities of daily living (ADL) and instrumental activities of daily living (IADL). Many studies reported a strong association between functional impairment and UI. For instance, a cross-sectional study done by Wang et al. [21] showed that subjects with UI had poorer physical function.
It is important to consider a patient's drug history, particularly in new-onset incontinence and in the elderly, where polypharmacy is common [37]. Wang et al. [21] showed that subjects with UI had more polypharmacy.
e current study showed a significant relationship between UI and laxative use (P value � 0.043). Similar result was reported by Blekken et al. [41] who studied 261 patients and concluded that UI was associated with constipation and laxative use.
Comorbidities are common in the older adult population, and UI can be caused by, associated with, or worsened by these. One study found UI to be independently associated with having at least one geriatric condition in 60% of study participants, at least two in 29% and at least three in 13% [42]. In the current study, UI was significantly associated with osteoarthritis, stroke, and vaginal prolapse.
is agrees with Turner-Stokes and Frank [43] proved that arthritis is associated with impaired mobility and dexterity and can lead to functional UI. Shaw and Wagg [42] concluded that UI is commonly associated with neurological conditions, including stroke, dementia, and Parkinson's disease. In the same way, Pandeva et al. [44] reported on 152 female participants in a tertiary referral center. e pelvic organ prolapse symptom of a vaginal bulge was present in 11% (17/152). Among cases of vaginal prolapse, 59% were accompanied by UI.

Conclusion
Urinary incontinence is prevalent in frail elderly females. Quality of life is significantly impaired in patients with UI, especially with mixed type and longer duration of symptoms. UI is significantly associated with older age, functional impairment, multiparity, osteoarthritis, stroke, vaginal prolapse, and laxative use.

Strengths of is Study.
is study has several strengths. First, it confirmed a high prevalence of UI (an important geriatric syndrome) with the application of multiple assessment tools in Arabic language in frail elderly females, a population relatively less included in previous research studies in our country. Second, we analyzed a wide range of risk factors/comorbidities in relation to UI in frail elderly females. ird, the study findings add to the growing body of literature on predictors and risk factors of UI in frail elderly Egyptian subjects.

Limitations of is Study.
Our study has some limitations. e relatively small sample sizes make generalizability of study findings cautious. e use of a cross-sectional design did not allow comparison with nonfrail subjects.

Data Availability
All the data used to support the findings of this study are included within the article and any further data can be provided from the corresponding author upon reasonable request.