Urinary incontinence (UI) is a common disorder with a substantial impact on quality of life (QoL). UI is a truly debilitating from a physical, psychological and social point of view and  may cause both isolation and loss of independence. UI affects up to 11% of men aged 60–64 years and 30% of men ≥85 years [1].  Should treatment not result in symptom control, then containment of incontinence with continence products is an option.  Whereas the wearing of continence pads in women with bladder problems is common and perhaps more accepted, that in men is less so [2]. Whilst it may be assumed that use of pads is an effective means to achieve social continence, there is little evidence as to the extent pad use affects quality of life. This study aimed to investigate the effect on quality of life of pad use in men with mild to moderate urinary incontinence in order to assess the feasibility of conducting a fully powered trial.

This was a quasi-experimental prospective cohort pilot study in community dwelling adult men (≥18years) who had not previously used containment products. Recruitment efforts and field notes were tracked to gain an understanding of feasibility of a future planned study.  Burden of questionnaire assessments was assessed. Informed, written consent to participate was gained from each man prior to any study procedure. Each man completed a baseline series of symptom and quality of life questionnaires, then provided with a supply of continence pads (2 pads/ day) for 6 weeks. After six weeks, the quality of life measures were repeated.
Measures included at baseline and 6 weeks , International Consultation on Incontinence Lower Urinary Tract Symptoms Quality of Life questionnaire (ICIQ-LUTSQoL), EuroQoL -5Dimension generic quality of life measure (EQ-5D),  International Consultation on Incontinence Male Lower Urinary Tract Symptoms Short Form questionnaire (ICIQ- MLUTS-SF), International Prostate Symptom Score (IPSS), Sandvik index, at baseline, duration of incontinence and demographics were collected.
The change in primary outcome measure, ICIQ-LUTSQoL and secondary outcome, EQ-5D total visual analogue score between baseline and end of study was used to calculate an effect size from which a power calculation was performed (G-power). This will inform a fully powered trial.
As there were no data upon which to power this exploratory study, the mean change and variance in ICIQ- LUTSQoL from a trial of treatment in women was used as a proxy, resulting in a desired sample size of 23 [3]. Analysis was by Student’s t test for matched pairs
Men were not offered any incentive to participate.  Travel expenses related to study visits, and a free supply of containment products were provided.

Recruitment:  Recruitment efforts were varied, including internet and local radio advertisements, talks to local community men's health groups, Prostate Cancer support groups, posters and washroom door notices around the city, main hospital outpatients and university campus, identification of potential participants at urology and continence clinics. No route was  more successful than another.  Over 6 months, 46 men were screened and 17 recruited to the study. Reasons for screen failure included varying previous  pad use and requests for payment to participate. Three men dropped out prior to completion, 1 due to travel, and 2 due to pad incompatibility. The mean (SD) age of the included men was 69.1 (14.8) years (range 33 -  89 years).  The distribution of the duration of incontinence prior to inclusion was:  ≤ 1 year; n=3,  1-2 years; n=2, 2-5 years; n=2,  > 5 years; n=7.  The questionnaires appeared feasible, taking an average of 20 minutes per participant to complete. Baseline Sandvik index indicated moderate severity and total IPSS was 15.1.  Table 1 shows the baseline and end of study outcome indicators.  Of all ICIQ-LUTSQoL subdomains, only "overall interference" (n/10) showed a statistically significant shift over the course of the study (mean (SD) 5.6 (3.3) baseline v 4.4 (3.2) end of study, p=0.028)
Using the change in total ICIQ-LUTSQoL and total EQ-5D VAS to calculate an effect size gives an effect size of 0.1 and 0.33 respectively.  The required sample size for a fully powered study with power of 0.8 and alpha of 0.05 is n= 740 using total ICIQ- LUTSQoL and n=76 using total EQ-5D VAS.

Recruitment proved difficult and labour intensive with a high screen failure.  Men were reluctant to come forward and the study failed to reach its intended sample size, although a robust estimate of variance on the primary endpoints was probably achieved.  Participants tended to have a long duration of incontinence.  The questionnaires were not perceived as burdensome upon the recruited men who mostly had moderate degrees of incontinence.  An effect size using total VAS in EQ-5D appears feasible to conduct a fully powered study. ICIQ-LUTSQoL hardly changed in response to pad usage suggesting either an insensitivity to change with a containment intervention in men, or a failure to measure the things that mattered to men. A qualitative study of men's use of pads, currently lacking in the literature, is planned to further explore this area.

The ICIQ- LUTS QoL questionnaire is insensitive to change in an exploratory study of the effect of pad use on the quality of life of men naïve to pad use.  A trial using a generic QoL scale such as EQ-5D appears feasible.  Recruitment is difficult and requires multiple avenues to achieve.

J Urol, 2006. 176(5): p. 2103-8; discussion 2108.
Int J Clin Pract, 2009. 63(6): p. 869-73.
Neurourol Urodyn. 2015;34(8):747-751.