Falls, a critical geriatric syndrome, affect over 14 million, or 1 in 4 older adults. The prevalence of urgency urinary incontinence (UUI) in older adults is notably high, contributing to falls, particularly during toileting-related activities. (1) With 36% of women aged 65 and above affected by UUI it is important to study whether measures of physical function known to predict a higher risk of falls differ in those with and without UUI. Previous studies have shown slower gait to be associated with higher risk of falls, decreased survival and higher disability. (2) The five-time sit-to-stand test and static balance assessments have also been shown to indicate fall risk. Thus, this study aims to analyze the impact of urgency urinary incontinence (UUI) on these gait, strength, and balance measures.

A two-group retrospective cross-sectional study analyzed gait parameter data of healthy community-dwelling ambulatory women, over the age of 60 with and without UUI. To be included in the UUI group, our criteria were that women reported incontinence at least twice weekly for >= 3 months despite correction of previously reversible cause which was confirmed by 3-day voiding diary with 24-hour pad test. Age, weight, height, and 3-day bladder diaries were collected. We computed participant’s body-mass-index (BMI) using height and weight and the mean leaks over 3 days. Medical history including polypharmacy, neurological, cardiovascular, musculoskeletal and diabetes mellitus conditions was recorded. URIS-24, CES-D and Beck anxiety questionnaires were administered to assess the quality-of-life impact of incontinence, depression, and anxiety respectively. Additionally, participants completed balance (narrow, tandem, and unipedal stance time), strength (five times sit-to-stand test (STS)), and gait speed testing, for which time taken to walk 4 meters at participants usual pace was recorded. Mean of two attempts at 5 times STS and gait speed were used for statistical analysis. Descriptive statistics, two-sample t-tests for independent variables and linear and multivariate regression for the dependent variables including balance, strength and gait measures were performed using R studio v2022.12.0-353.

Fifty-two women were included in the study. The study sample had a mean ± standard deviation age of 67.5 ± 6.9 years, gait speed of 0.8 ± 0.5 m/s, and 5 times sit-to-stand (STS) time of 12.9 ± 4.8 seconds. Eighteen participants (34%) in the study were continent and thirty-four participants (65%) were incontinent (age: 64±4.9 yrs. vs 69±7.4 yrs., p=0.02). Participants with incontinence had a significantly higher BMI (31.7±6.5 kg/m2 vs. 26.9±4.6 kg/m2, p=0.00) and reported a significantly higher CES-D score (7.1 ± 5.1 vs 3.6 ± 2.9, p=0.01), Beck anxiety score (8.3 ± 5.7 vs 4.1 ± 3.5, p=0.00) and lower URIS-24 score (70.5 ± 19.1 vs 115.8 ± 6.6, p=0.00). The number of participants with polypharmacy (five or more prescription drugs) was comparable between the two groups as was the incidence of neurological, cardiovascular, Musculoskeletal, and diabetes mellitus conditions. On univariate regression gait speed was significantly slower with higher frequency of leaks (F (1,50) =8.6, p=0.00, R2=0.14, with leak frequency β=0.07, t=-2.9, p=0.00). After controlling for age and BMI on multivariate regression the relationship persisted (F (3, 48) =5.147, p=0.00, R2=0.24, with BMI β=-0.02, t=-2.05, p=0.04 and leak frequency β =-0.05, t=-2.17, p=0.03). Participants with higher UUI leaks had significantly shorter unilateral stance time on linear regression (Right stance: p=0.02; Left stance: p=0.00) which was significantly influenced by only age and BMI on multivariate regression. The number of leaks was not associated with 5-times STS time, eyes closed narrow stance time, or tandem stance time.

The analysis showed that continent and incontinent groups were significantly different in age and BMI, both of which have an impact on gait and falls. Thus, we conducted a regression analysis controlling for age and BMI to account for this discrepancy. Even after controlling, the continent group has faster gait speed compared to incontinent group within which, gait speed is significantly related to number of leakage episodes. This study cannot interpret direct causality, but UUI has previously been proven causative for many factors including distracted mind and interrupted sleep which may affect gait speed. (3) Conversely gait speed may affect ability to reach the toilet in time increasing UUI leakage episodes. Further studies of these complex intertwined factors are necessary.

Gait speed is slower in individuals with higher frequency of leakage with UUI and therefore may contribute to the increased risk of falls. Our findings suggest that gait speed may be more sensitive to differences in physical function associated with UUI than 5-times STS or balance measures. Gait speed may be an important variable to target screening and assessment and inform treatment of those with UUI at higher risk for falls.

Decalf V, Bower W, Rose G, et al. Prevalence and characteristics of incident falls related to nocturnal toileting in hospitalized patients. Acta Clin Belg. 2021;76(2):85-90. doi:10.1080/17843286.2019.1660022
Studenski S, Perera S, Patel K, et al. Gait speed and survival in older adults. JAMA. 2011;305(1):50-58. doi:10.1001/jama.2010.1923
Gibson W, Jones A, Hunter K, Wagg A. Urinary urgency acts as a source of divided attention leading to changes in gait in older adults with overactive bladder. PLOS ONE. 2021;16(10):e0257506. doi:10.1371/journal.pone.0257506

Continence 12S (2024) 101633
DOI: 10.1016/j.cont.2024.101633