There is a strong but unexplained association between lower urinary tract symptoms (LUTS), overactive bladder (OAB) and falls in older men and women, with reported odds ratios for falls in those with nocturia, urgency, and incontinence between 1.5 and 2.3 [1]. Little attention has been paid to the potential causes of this association and this has variably been suggested as being due to rushing to get to the toilet, slipping in urine following an episode of incontinence, or there being no conceivable causal relationship. However, in continent, middle-aged women a strong desire to void induces changes in gait including a slowing of gait speed and reduction in stride length [2]. The control of continence is a cognitive task with multiple areas of the brain being involved in the control of the lower urinary tract [3]. Diverted attention is the concept whereby the simultaneous execution of two cognitively demanding tasks leads to decline in performance of both tasks. It is well recognised that diverted attention causes changes in gait in older adults.
We hypothesised that urinary urgency acts as a source of diverted attention in older adults with overactive bladder (OAB), and that this induces gait changes which predispose to falls.

Participants were recruited from a specialist continence clinic and through local newsletters. Inclusion criteria were: age 65 or older, a clinical diagnosis of OAB meeting the ICS definition. Participants had a daytime frequency of ≥8 and at least weekly urgency incontinence. Exclusion criteria were: cognitive impairment, defined as a Montreal Cognitive Assessment score of 25/30 or lower, neurological disease that would interfere with cognitive function or gait, such as stroke or Parkinson’s disease, and the inability to walk 30 metres unaided for any reason.
Participants underwent 3D motion gait analysis while walking for 30 metres under three conditions; undistracted and with an empty bladder, distracted by performing the n-back test, a validated source of diverted attention and when experiencing self reported, ICS defined urinary urgency. Urgency was induced by asking participants to drink non-caffeinated fluids ad libitum until they experienced an urgent need to void that they could not defer. Distraction was induced by the simultaneous performance of an auditory n-back test, in which a series of letters is read to the participant, and if the letter is the same as one two prior in the sequence they indicated this verbally to the tester.
To record gait parameters (velocity, cadence, and stride length) reflective markers were applied in a, with markers on the forefoot, heel, anterior and posterior trunk. A series of video cameras recorded the position of each marker as the participant walked, and from these data velocity, cadence, and stride length were calculated. The Berg Balance Score and Activity-Based Confidence Scre was also recorded.
In order to ensure participants were able to defer voiding for as long as possible, the gait laboratory was located near to a toilet facility reserved for the use of the participant during data collection, and the final walk was towards that toilet. 
Each parameter was then compared under each of the three conditions using a repeated-measures ANOVA with Bonferroni correction. Statistical significance was set at p<0.05. Based on previous wokd with gait under dual-task condtions, we calculated a sample size for a within-subject design of 10. However, as the effect of urgency and diverted attention has not previoulsy been studied, we aimed to over-recruit by a factor of 2.

27 participants were recruited, of whom 22 were women (81%). The mean age was 75.3 (SD 5.8) years. The average Berg Balance Score was 46 (out of 56) and the average Activity Based Confidence Score was 83.2%, indicating that these participants were at low risk of falls. The demographic, cognitive, and balance data are summarised in Table 1.

Urinary urgency and distraction both induced statistically significant changes in gait, with a reduction in both velocity and stride length. Velocity reduced from 1.1m/s when undistracted to 1m/s (p=0.08) when experiencing urgency and 0.8m/s (p<0.001) with distraction. Stride length reduced from 1.2m when undistracted to 1.1m with urgency and 1m when distracted, both p<0.001. Cadence reduced with distraction, from 110 steps/min to 94 steps/min (p<0.001) but was unnafected by urgency (108 steps/minute, p=0.8). The gait data are summarised in Table 2.

This is the first study to directly compare gait in older adults experiencing urinary urgency and distraction. In this mobile, community dwelling sample of older adults with OAB, the sensation of urgency induced similar changes in gait to distraction, and those gait changes, reduced velocity and reduced stride length, have both been shown to increase falls risk in older adults. These data suggest that urinary urgency may act as a source of diverted attention and that this is, at least in part, the explanation for the observed association between LUTS and falls in older adults.

Urgency induced changes in gait that were similar to those induced by distraction. These changes have been associated with an increased falls risk in older adults. These results suggest that urgency may act as a source of diverted attention in older adults with OAB. Given that dual-task training, a form of physiotherapy in which gait and balance work combines a simultaneous cognitive task, has been shown to reduce the effect of distraction on gait, an intervention trial of dual-task training in adults with OAB should be performed.

Gibson, W., et al., The association between lower urinary tract symptoms and falls: Forming a theoretical model for a research agenda. Neurourol Urodyn, 2018. 37(1): p. 501-509.
Booth, J., et al., The Association between Overactive Bladder and Falls in Older Women: A Proposed Explanation. Neurourology and Urodynamics, 2013. 32(6): p. 805-806
Fowler, C.J. and D.J. Griffiths, A decade of functional brain imaging applied to bladder control. Journal of Urology, 2011. 186(2): p. 615-616.