Faecal incontinence in rural and regional northern Queensland community-dwelling adults

Introduction: In Australia, faecal incontinence, the involuntary loss of liquid or solid stool with or without a person’s awareness, has been reported in 8% of the South Australian and 11% of the urban New South Wales community-dwelling populations. Studies conducted in 2004 and 2005 reported faecal incontinence in more than 20% of colorectal and urogynaecological clinic patients at Townsville Hospital (a referral centre serving rural North Queensland). This prompted concern regarding the level of faecal incontinence in the community. The aim of this study was to investigate the prevalence of faecal incontinence in the North and Far North Queensland urban and rural communities. Methods: The sample size was based on the New South Wales postal surveys (11% prevalence). Higher rates were expected in North/Far North Queensland, so prevalence there was estimated at 12.1% (confidence interval ± 2%, ie the true level to be between 10.1% and 14.1%). The sample for each of the Townsville, Cairns (in Far North Queensland) and rural/remote settings was calculated at 1022. The database for the present study was compiled using a systematic randomised process selecting two private names from each column on each page of the Cairns and Townsville White Pages® (Cairns: 1112 urban, 481 rural, 226 remote; Townsville: 1049 urban, 432 rural, 320 remote). The questionnaire covered personal demographics, health/risk factors, bowel habits, nutrition (fibre and fluid intake) and physical activity. Faecal incontinence was defined as accidental leakage of solid or liquid stool in the past 12 months that was not caused by a virus, medication or contaminated food. To improve the response rate a participation incentive of a chance to win a $250 voucher or one of ten $50 vouchers was offered in the initial mail-out. The initial survey was mailed out in July 2007; two follow-up surveys were mailed out to non-responders in September 2007 and January 2008. One hundred randomly selected non-responders were telephoned in February 2008. Results: A total of 1523 responses provided a 48.1% response rate. Faecal incontinence prevalence was 12.7% (174/1366) with no gender or locality differences. Prevalence increased significantly with age in men (p=0.034), but not in women. Only 10 respondents with faecal incontinence consulted their doctor in the previous year for this reason. Incontinent respondents had


Introduction
Accidental bowel leakage is rarely disclosed 1 .It can range from occasional loss of flatus, through staining of underwear with mucus or stool, to faecal incontinence, which the International Continence Society defines as the involuntary loss of liquid or solid stool that is a social or hygienic problem 2 .The impact of faecal incontinence on quality of life can be debilitating and embarrassing.Stringent coping strategies often alienate the sufferer from friends and family 3 .There is little awareness of it in the general community.Even sufferers confuse it with diarrhoea, faecal urgency, irritable bowel syndrome or inflammatory bowel disease 4 .Risk factors resulting in the development of faecal incontinence include congenital anorectal abnormalities; neurological or spinal damage; obstetric or anal trauma; anal or rectal cancers; inflammatory bowel disease; reconstructive bowel surgery; psychological problems; abdominal/pelvic irradiation; infections; reactions to medications, drugs or diet; rectal prolapse, anal fistula or haemorrhoids; immobility; increasing age; chronic constipation; obesity and poor management of diarrhoea or loose stool; and idiopathic causes 5-7 .
Internationally, the prevalence of faecal incontinence in community-dwelling adults ranges from 0% to 15.2% 8 .Australian studies have estimated prevalence between 8% 9 and 11% 10,11 in community-dwelling adults, with telephone interviews being used in a South Australian study 9 and two postal surveys being conducted in Sydney, New South Wales 10, 11 .Studies undertaken at the Townsville Hospital urogynaecology and colorectal outpatient clinics reported up to 26% of patients having the condition, with a considerable effect on their quality of life 3,12, 13 .
Currently up to 72% of nursing home residents suffer with faecal incontinence 14 .The proportion of Australians over the age of 65 is expected to increase from 13.4% of the population in 2007 to 25.3% of the population by 2047 and the percentage of those over the age of 85 will likely rise from 1.7% to 5.6% 15 .Many Australians wish to age in their homes rather than move into specialised care 16 .Increased demand for community-based incontinence services will be due to not only the ageing population 15 but also the fact that younger cohorts (baby boomers, X and Y generations) are more assertive and expect to have their needs satisfied 17 .World War II (age 85-90 in 2012) and post-war cohorts (67-84) are known as the silent generation and are stoic and reserved, respect clinicians and are more likely to comply with their recommendations.This general attitude may be reflected in their reticence to disclose embarrassing issues such as faecal incontinence and reluctance to seek treatment for this condition.
Baby boomers (47-66) are sceptical, questioning and expect pharmacological solutions to their age-related problems 17 .
They are more likely to try novel treatments such as anal implants 18 , sacral nerve or percutaneous tibial nerve stimulation 19 for their faecal incontinence, with the hope of finding a 'magic pill' or panacea 17 .With compulsory superannuation they are prepared to pay for healthcare services into their old age 15 .Generation X (36-46) treasure independence, are focused on their quality of life and work to fund their more balanced lifestyle.Their choice of health service providers are likely to be an economic decision and they will require more personalised support, shared decisionmaking and greater self-management 20 .The Y generation (18-36) are tolerant team players who value mentorship and have grown up with the internet and frequent technological change.They value excitement and instant gratification through email, messaging and social media and may adopt healthcare services via social media 17 ; they may prefer selfmanagement of faecal incontinence using telephone and tablet applications.For this and other age-related health problems the Ottawa Charter action areas of building healthy public policy, creating supportive environments, strengthening community action, developing personal skills and reorienting health services provide a useful framework to develop prevention, early intervention and treatment protocols 21 .
This study was undertaken to investigate personal demographics, health/risk factors, bowel habits, nutrition intake (fibre and fluid) and physical activity in relation to faecal incontinence.Here the authors report the prevalence of faecal incontinence in community-dwelling adults and assess disclosure of accidental bowel leakage in regional and rural areas of North/Far North Queensland, Australia.

Methods
A sample of 3620 subjects was compiled from the 2006-2007 Cairns and Townsville White Pages® telephone directories (Cairns: 1112 urban, 481 rural, 226 remote; Townsville: 1049 urban, 432 rural, 320 remote) using systematic random sampling.The first non-business name and address was selected from each column between 90-100 mm and 190-200 mm from the top of the page.The sample addresses were updated from the 2007-2008 printed and online White Pages® directory prior to distribution.Where a subject no longer appeared in the directory the next alphabetically listed private address was chosen as a replacement.
The self-administered questionnaire (Appendix 1), a study information sheet, an incentive leaflet (return of the completed survey provided an opportunity to win a $250 or $50 voucher), an opaque incentive response envelope, and a reply-paid envelope were mailed to everyone on the database.The survey contained 62 questions in five sections including personal demographics (11) with additional questions about female obstetric history (8), health/risk factors (6), bowel habits (32), nutrition (fibre and fluid intake) (2) and physical activity (3).
Demographic questions included those related to age, gender, cohabitation, residence type, postcode, education level, country of birth, height, weight and indigenous status.Female obstetric history questions included menopausal status, number of natural and caesarean births, use of forceps/vacuum, episiotomy, post-tear stitches, hysterectomy and hormone replacement therapy.Health or risk factor questions included participant-perceived general health as well as factors, previously identified, as potentially causal for this condition 5-7 (Appendix 1).
Participants were asked for information about year and results of a colonoscopy.Questions about bowel habits included those related to frequency of defecation; stool type 22 ; urgency; difficulty in emptying bowels; constipation and straining; accidental anal leakage and its effect on quality of life; and coping strategies used.Nutrition questions related to fluid and fibre intake.Activity questions related to normal daily activity, exercise and pelvic floor exercises.
The initial survey was mailed out in July 2007, followed by two follow-up surveys mailed to non-responders in September 2007 and January 2008.One hundred random non-responders were invited to answer the survey by phone in February 2008.All mailouts contained reply-paid envelopes and covering letters explaining the purpose of the study.The follow-up mail-outs included a non-response/remove me from the database tool seeking reasons for non-response.Anonymity of replies was maintained.Core faecal incontinence questions were adapted from the Cleveland Clinic Florida Fecal Incontinence Score (Wexner) which correlates well with clinical presentation of faecal incontinence and has been referred to as 'a tool of choice' for assessing faecal incontinence in community-dwelling older adults 23 .
In this study, faecal incontinence was defined as accidental leakage of solid or liquid stool in the previous 12 months that was not caused by a virus, medication or contaminated food.

Statistics
Community prevalence and 95% confidence intervals (CI) of faecal incontinence and accidental bowel leakage were determined.Numerical data are given as mean value and standard deviation or median value and interquartile range (IQR), depending on the distribution.Comparisons between characteristics were conducted using χ 2 tests and χ 2 tests for trend, nonparametric Wilcoxon tests, and student's t-tests.Statistical analyses were conducted using Statistical Package for the Social Sciences Windows v17 (SPSS Inc., http:// www.spss.com).Throughout the analysis p<0.05 was considered statistically significant.The sample size calculation was based on the Sydney, New South Wales community postal surveys (faecal incontinence prevalence = 11%) 10, 11 .The authors expected the North/Far North Queensland prevalence to be greater.Choosing 12.1% (10% higher) with a confidence level of ± 2% (ie true level to be between 10.1% and 14.1%), the sample size calculation for each of the Townsville, Cairns and rural settings was 1022.Systematic random sampling generated 3620 private names for the database.

Ethics approval
James Cook University Human Research Ethics Committee granted ethical approval (H2630).

Results
Of the 3620 surveys mailed out, 432 were returned to sender as 'no longer at this address' (18 of these were reported deceased); nine completed surveys were invalid (one underage, two nursing home residents, six living outside North/Far North Queensland).Fourteen of the 100 random numbers telephoned were inactive phone numbers.Of the remaining 3165 potential participants, 95 gave reasons for not wishing to participate; 1523 questionnaires were returned giving a response rate of 48.1%.Although more women (11.3%) than men (7.8%) resided alone (p<0.001),there were no significant differences between the genders with regard to type or location of residence, level of education, occupation type, country of birth or indigenous status.
Of those with faecal incontinence, 54 (34.8%) reported urge incontinence, 19 (12.3%) passive incontinence and 44 (28.4%) both passive and urge incontinence.Only 28.1% of incontinent participants could always differentiate between flatus and stool compared with 53.5% of continent respondents (p<0.001,Table 2).Nocturnal bowel leakage was a problem for 41 (26.3%) of incontinent respondents.While 59 (36.6%) incontinent respondents had problems modifying their diet to control their bowel, 49 (32.7%) did not know which medications would control their bowels and 56 (34.1%) felt they had no control over their bowels; only 11 (7.1%) had sought treatment for this condition.More than 38% (64) of respondents with faecal incontinence reported having first degree relatives with 'bowel problems'.
A comparison of respondents with and without incontinence showed that more with incontinence reported poor or very poor general health; more frequent, irregular, incomplete/fragmented defecation, and looser stools; inability to delay by 15 minutes or requiring aids to defecate; alternating diarrhoea and constipation; a history of constipation; and pad-wearing (all p<0.001,Table 2).In addition, more incontinent respondents reported that their bowel function negatively affected their daily activities such as work, sports, housework/gardening, social activities, travel, relationships and sex life (all p<0.001).More than half the incontinent respondents (82/162) reported difficulty finding public toilets.
Embedded within the survey were a number of questions relating to accidental bowel leakage in addition to the Wexner score 5 .Table 1 presents the disclosure of accidental bowel leakage with the most severe element counted for each respondent.Those who did not report faecal incontinence, but disclosed accidental bowel leakage due to an acute illness, are reported separately in Table 1 (n=245, 97 male) even if they also reported other accidental bowel leakage items.Stool-related accidental bowel leakage (18.2%) included faecal incontinence, soiling with flatus (not including possible mucus-related soiling with flatus, n=3, one male), and soiling with urgency.This was not statistically different for gender or location/rurality of respondents' residences but was significantly higher for the over-40 age groups (p=0.040).Possible stool-related accidental bowel leakage (3.0%) included leakage of mucus; staining underwear, if it was considered a problem in the month prior to completing the survey; accidental bowel leakage that was found bothersome; and passive accidental bowel leakage.Of the remaining 1109/1513 respondents who completed these questions 346 (22.9%) reported they could not always differentiate between flatus and stool, 69 (4.6%) reported accidental leakage of flatus and only 532 (35.2%) respondents reported no concerns about accidental bowel leakage, and no accidental bowel leakage.
Types of accidental anal leakage are presented in Table 4 and did not differ across regions.Leakage rates of solid and liquid stool, mucus and accidental flatus were similar for both genders although staining of underwear was significantly higher in men (26.8%) than in women (17.3%, p<0.001) and for older respondents (<39: 15.9%; 40-59: 28.8%; >60: 33.7%, p<0.001).

Discussion
The main finding of this study was that 12.7% of adult North/Far North Queensland community members reported faecal incontinence in the previous 12 months that was not due to an acute illness.This rate increased with age for men although, in total, there were no gender or locality differences.When soiling with flatus and urgency were included, stool-related accidental bowel leakage was substantially higher, at 18.2%.
The prevalence of faecal incontinence is highly dependent on the definition applied 10, 24 .This study's focused definition -'accidental loss of solid and/or liquid stool in the previous twelve months, not due to a virus, medication or contaminated food' -has not, to the authors' knowledge, been used previously.Despite using this refined definition the rate of faecal incontinence in this study is higher than the two earlier Australian postal surveys 10,11 although the severity (type and frequency) was similar.Using a broader definition of accidental stool leakage that did not exclude faecal incontinence resulting from an acute illness, the prevalence was much higher, at 28.1% (Table 3).Thus previous concerns of high levels of faecal incontinence in the North/Far North Queensland community are warranted  #).¶ Each respondent is counted once only (worst severity noted).Ten missing -respondents did not answer any of the questions about disclosure of accidental bowel leakage.§ Faecal incontinence (accidental solid and/or liquid leakage) in past 12 months not due to an acute illness; prevalence (95% confidence interval=12.7%(10.9-14.5%))was based on response to Cleveland Clinic Florida-Fecal Incontinence Score (Wexner) questions only (n=1366, 572 men, 794 women).‡ Does not include possible mucus-related soiling with flatus (three: one man, two women). In participants who have not disclosed faecal loss (solid/liquid/mucus) but whose staining of underwear was a problem in the month prior to survey completion.# Participants who did not report faecal incontinence but reported accidental bowel leakage due to virus, medication or contaminated food included here whether or not they also reported any of the other stool, possible stool or flatus-related items.§ Cleveland Clinic Florida -Fecal Incontinence Score.‡ In participants who have not disclosed faecal loss (solid/liquid/mucus) but who currently soil clothing while passing wind and/or where staining of underwear was a problem in the month prior to survey completion.
One strength of this survey was that it enabled respondents to disclose any type of accidental bowel leakage, whether they termed it faecal incontinence or not.In a recent New Zealand study, researchers combined three different measures -a bowel control problem, quality of life impairment and faecal incontinence ≥1/month -to better determine faecal incontinence prevalence in the community 25 .While the researchers reported 12.4% of respondents leaked solid or liquid stool at least once per month and 26.8% had impaired quality of life due to accidental bowel leakage, they calculated community prevalence at 13.2% using the overlap of at least two of three measures to redefine faecal incontinence.The aim of the present study too was to provide a more accurate estimate of those suffering with faecal incontinence.In this study, issues such as flatus and mucus were excluded.The New Zealand survey tool may have included these in the quality of life component 26 , thereby overstating the prevalence.Up to 50% of studies previously reviewed 8,24 calculated prevalence using faecal incontinence occurring in the previous 12 months as done in the present study.As in those studies, the rate in this study may be higher than that of the New Zealand study and others using incontinence reported over shorter time frames.A troubling result is that only of the respondents had no bowel concerns.Further, many respondents with faecal incontinence felt they had no control over their bowels and didn't know how to manage their diet or which medications to take to mitigate the problem.As only 7% actually sought medical advice, fear, embarrassment and the stigma associated with faecal incontinence continues to hamper disclosure as previously identified 27 .
As more people age 'in place' 16 the prevalence of faecal incontinence in the community is likely to increase toward that of aged-care facilities.Younger cohorts, which are less stoic, are unlikely to tolerate this condition and will demand treatment 17,20 .Ageing populations increase healthcare costs 15 , and increased demand for faecal incontinence treatment will further escalate these costs.The concepts of the Ottawa Charter for Health Promotion 21 can be used to help reduce this added financial burden by encouraging prevention of constipation and obstetric damage, and ensuring benign clinical diagnoses are managed with conservative programs (potentially self-managed) including dietary management, pelvic floor exercises, defecation and relaxation techniques, and home biofeedback 28 .
Limitations of this study include a low response rate of 48.1% despite mailing two follow-up questionnaires.The average age of survey respondents was 54 years, 9 years older than the North/Far North Queensland adult population (45 years), which is marginally younger than the state (46 years) and national (46.5 years) averages 29 .Reasons for the low response rate could include lack of interest in the subject matter, particularly among younger survey recipients.Furthermore, younger people may be under-represented, not only because they may have chosen not to respond to the questionnaire, but also because they are more likely to only have a mobile telephone and thus not have received a survey because they were not listed in the telephone directory 30 .Older respondents may have self-selected due to a particular interest in the topic or because of a personal bowel issue 24 .Townsville, the largest city in North Queensland, has a highly mobile population with a 35% net transient population measure 31 .Of the sample of 100 non-respondents telephoned in an attempt to further increase the response rate, 14 telephones had been disconnected.If this were representative of the population, it could also partially explain the low response rate.However, using the alternative data collection method of telephone interviews may have yielded an even lower response rate based on experience from the South Australian study 9 .Similar response rates using this or analogous study enrolment methods were found in a cross-sectional study investigating accidents and injuries in North Queensland 32 and a New Zealand faecal incontinence prevalence study using a random selection from the Canterbury electoral roll 33 .Thus the response rate may be a limitation of the method used to elicit information about a topic that is either 'sensitive' or does not interest a particular segment of the population.

Conclusions
There is a high level of untreated faecal incontinence in the North/Far North Queensland community which degrades the quality of life of those with this ailment.With a burgeoning ageing-in-place population the prevalence in this and other communities is likely to increase.Lifestyle education to prevent chronic constipation and encourage the development and maintenance of strong pelvic floor muscles among young people could reduce the prevalence of this condition.Faecal incontinence, its prevalence, prevention and treatments should be widely discussed to raise community awareness reducing the incidence and the destructive effect it has on the quality of so many lives.Eliminating the stigma of faecal incontinence should encourage those with the condition to overcome their embarrassment and seek early treatment.Where appropriate, conservative self-managed care is economically rational and can improve quality of life for all, not just those with access to the few specialist biofeedback clinics.

Figure 1 :
Figure 1: Proportion of respondents with faecal incontinence.

Figure 2 :
Figure 2: Frequency of anal leakage † in respondents with faecal incontinence

3,12,13 .Table 1 : Respondents' disclosure of accidental bowel leakage Accidental bowel leakage †
†In past 12 months not due to virus, medication or contaminated food (except acute illness related accidental bowel leakage -see note

Table 2 : Comparison of bowel habits and health in respondents with and without faecal incontinence Faecal incontinence † No faecal Incontinence † Bowel habits and health
Bristol Stool Form Scale; respondents who reported multiple or varied types of stool passed were removed (11 with, 25 without faecal incontinence).

Table 3 : Frequency of accidental faecal leakage in respondents †
Faecal incontinence (accidental solid and/or liquid leakage) in past 12 months not due to a virus, medication or spoiled food.§ Accidental loss of solid and/or liquid stool in the previous 12 months, due to a virus, medication or contaminated food.‡ Includes those who did and did not disclose faecal incontinence. Includes only those who did not disclose faecal incontinence. ¶

Table 4 : Accidental anal leakage in respondents †
† Anal leakage in 12 months prior to survey.¶ Prevalence and 95% confidence interval.
P. 2010 Survey of Health Care Consumers: key findings, Lewis SJ, Heaton KW.Stool form scale as a useful guide to intestinal transit time.Scandinavian Journal of Gastroenterology 1997; 32: 920-924.