Falls and urinary incontinence (UI) are both major issues affecting elderly women aged 65 and over. The fall rate in urge and mixed urinary incontinent elderly women is 29% compared to 20% in continent elderly women (1). However, the relation between falls and incontinence is still not well understood. One hypothesis is that a strong desire to void (SDV) could alter gait parameters and therefore increase the risk of falling (2). The primary objective of the study is to investigate the effect of a SDV on gait parameters in urge/mixed incontinent and continent community-dwelling women who are at risk of fall. The secondary objective is to determine the relationship between UI severity and gait parameters in the group of incontinent women.

An observational pilot study was undertaken with two groups of healthy community-dwelling women aged 65 and over, who experienced at least one fall in the past year with and without UI. To be included in the urinary incontinent group, a participant had to have moderate to severe urge/mixed UI as determined by the International Consultation on Incontinence Questionnaire on UI Short Form (ICIQ-UI SF) and >= 3 urine leakages/week in the 7-day bladder diary (with at least one urgency-related leak). To be included in the continent group, participants had to have an ICIQ-UI SF score equal to 0, no urine leakage reported in the past year and none in the 7-day bladder diary. Participants with a body mass index (BMI) >=35 and health conditions likely to influence gait or urinary continence during the study were excluded. After signing a consent form and completing a 24-h pad test, each participant participated in a 3-h gait laboratory assessment. Demographics and the results of a Montreal Cognitive Assessment test (MOCA) were acquired for all participants. History of falls was also recorded. After drinking water until they experienced a SDV as determined by a score of 3 on the Urinary Sensation Scale (USS), participants were asked to walk on a computerized gait analysis mat (GAITRite), on their way to the toilet. After emptying their bladder, i.e. with no desire to void (NDV), they were asked to walk again on the instrumented mat. Spatial and temporal gait parameters and their variability were calculated for the two groups. Descriptive statistics were obtained for demographic, cognitive, UI and gait data. Independent t-test and Chi-square test were used to compare the continent and UI groups for demographic, cognitive and UI outcomes.  An analysis of variance (ANOVA) with repeated measures was conducted to explore the differences between the two groups (continent/incontinent) for the two conditions (NDV and SDV). As BMI was significantly different between groups, we included BMI in our ANOVA  (BMI<25/>=25). In order to quantify the impact of the desire to void on gait, the Eta square effect size (η2) was also calculated for each of the gait parameters. Finally, for the incontinent group, the correlation between incontinence severity as determined by ICIQ-UI SF and gait parameters for both conditions (NDV and SDV) was computed using Spearman correlation tests.

Thirty-two women participated in the study; 17 continent and 15 urge/mixed urinary incontinent. Demographics, cognitive, number of falls and continence status/severity outcomes for each group are presented in Table 1. BMI, number of falls, ICIQ-UI SF scores, 24-h pad test weight, and number of urine leakages noted in the bladder diary were significantly different between groups favoring the continent group (Table 1). There was a statistically significant main effect between the two conditions with a large effect size for reduced velocity and stride width, and a moderate effect size for increased stance time and reduced stride length when experiencing a SDV (Table 2). However, there was no group effect. Of note, 6/15 (40%) of the incontinent women walked at a speed of under 100 cm/s in both conditions as opposed to only 2/17 (12%) of continent women with NDV and 3/17 (18%) with SDV. Furthermore, the stride length was shorter in women with a BMI >=25 (p=0.04) in both conditions. The BMI had no effect on other gait parameters in any condition. In the incontinent group, we found moderate correlations for the SDV condition with reduced velocity (rs= -0.45, p=0.01) and reduced stride length (rs= -0.49, p<0.01) related to more severe incontinence. In the NDV condition, reduced velocity (rs=-0.43, p=0.01), reduced stride length (rs= -0.49, p<0.01) and increased stance time variability (rs= 0.49, p<0.01) were moderately correlated with more severe incontinence.

Gait parameters were influenced by SDV, regardless of the group. The reduced velocity observed when experiencing a SDV was due to a shorter stride length and increased stance time.  We propose that it is easier to hold urine when experiencing SDV by reducing walking velocity and shortening stride width. Although they were no group effect, incontinent women showed a clinically slower walking velocity with NDV than their continent counterparts. When experiencing SDV, they reduced their already slower walking velocity. This is of importance because a walking speed under 100 cm/s is related to a higher risk of falls (3). In the incontinent group, UI severity was correlated with slower gait parameters and an increased gait variability. As increased variability is known to be related to a higher risk of falls, women with severe UI could be at higher risk (3).

To our knowledge, this is the first observational study of urinary incontinent and continent community-dwelling women who have experienced falls to report the influence of SDV on gait parameters. The  SDV affected the spatiotemporal parameters of gait, regardless of the continence status, which could increase the risk of falls. Finally, in those with more severe incontinence, gait parameters could be related to a higher risk of falls, in both conditions. More studies are needed to confirm these results and to further understand falls in an incontinent population.

Chiarelli PE, Mackenzie LA, Osmotherly PG. Urinary incontinence is associated with an increase in falls: a systematic review. Aust J Physiother. 2009;55(2):89-95.
Booth J, Paul L, Rafferty D, Macinnes C. The relationship between urinary bladder control and gait in women. Neurourol Urodyn. 2013;32(1):43-7.
Mortaza N, Abu Osman N, Mehdikhani N. Are spatio-temporal parameters of gait capable of distinguishing a faller from a non-faller elderly? European Journal Physical rehabilitation medecine. 2014;50:677-91.