Participants were consecutive patients attending the colorectal clinic (CC) and urogynecology clinic (UC) at The Townsville Hospital, in North Queensland Australia, between January and June 2003 and August and November 2004. Ethical approval was obtained from the ethics committees of Townsville Hospital and James Cook University.

The study procedure has been described previously[59]. Briefly, a cross-sectional survey was conducted among patients at the CC and UC. On arrival, all patients attending these clinics were invited to participate in the study. Exclusion criteria included children (under 18 years), pregnancy, terminal illness, mental illness, or gastrointestinal stoma. Eligible subjects completed a patient consent form. Participants were then given a self-administered questionnaire that contained questions about patient demographics, alcohol consumption, preexisting medical conditions, and prior surgical history; all known risk factors for FI. Patients who answered “yes” to the question “do you ever accidentally soil your clothes or underclothes with feces?” proceeded to questions relating to the frequency, severity and management of FI, and the 29 question FIQL survey tool[22]. The physicians of participating patients also surveyed them using the Cleveland Clinic Florida Fecal Incontinence Score[25], which was compared with the self-administered survey tool results to investigate disclosure problems[9]. Here, we report information about the QOL questions from the self-administered questionnaire.

Mean and SD were used to describe age. Medians and interquartile ranges were used as measures of central tendency and estimates for dispersion for duration of FI and FIQL scales. FIQL scales were calculated in accordance with the developers recommendations[26]. For bivariate testing of categorical variables, exact versions of χ² tests were conducted for nominal items, while exact trend tests were used for ordinal variables. Multiple linear regressions were used to investigate relationships between FIQL scales and clinics, basic severity (type × frequency) and duration of FI among female participants, and FIQL scales and components of severity[2], with and without quantity of fecal loss, in all participants. Potential components were initially considered separately and were then combined. As FIQL scales were not distributed normally, they were transformed by taking the square root. As no major differences between transformed and untransformed regression results were found, the untransformed regression data have been presented. Results of linear regression analyses are presented by regression coefficients (r) with 95% CIs and/or P values. A significance level of 0.05 was adopted a priori.

A total of 769 patients (451 in 2003 and 318 in 2004) were invited to participate. Of these, 423 women attended the UC and 146 men and 200 women attended the CC, with 93.4% (n = 718) of the eligible patients completing the self-administered FI questionnaire. As 43 patients completed the survey more than once, the initial responses were used and duplicate data were removed from the combined database, which left a total of 675 unique entries.

Of the 675 patients in the study 154 (22.8%; 95% CI: 19.6%-26.0%) reported having accidentally soiled their clothes or underclothes with feces and answered the FIQL questions. There were 27 men from the CC, and 127 women, 52 from the CC and 75 from the UC, with FI. The mean age (SD) of the participants with FI was 56.2 (14.3) years. There was no age difference between sexes (P = 0.281) or clinics (P = 0.82), or the women attending the two clinics (P = 0.87).

Patients reported the following etiological risk factors for FI. There were 27 participants who reported having bowel disease (25 from CC, nine male), with 14 of them (all from CC, two male) having been diagnosed with colorectal cancer. Twenty seven women (eight from CC) had undergone vaginal repair surgery, whilst 51 women (16 from CC) reported difficult vaginal births using forceps, vacuum extraction or long second-stage labor. Twenty-one participants (11 from CC, two male) reported rectal prolapse; 23 (20 from CC, nine male) had undergone surgery for hemorrhoids, anal fissures or fistulas; 101 (39 from CC, 10 male) had urinary incontinence; 11 (six from CC, four male) had anal injuries; 18 (10 from CC, four male) had diabetes; six (four from CC, two male) reported spinal cord disease; three (one from CC, one male) had neurological disease; 39 (21 from CC, seven male) reported chronic constipation; and 17 women (five from CC) had psychiatric problems that required medication. Some patients had multiple factors.

Patients attending the UC documented having FI for a significantly longer period (median: 24 mo, IQR: 12-60), than those attending the CC [12 (6-24), P = 0.001). Similarly, women in the UC clinic had FI for longer than those in the CC clinic [UC = 24 (12-60); CC = 12 (6-24), P = 0.002].

More than half (55.8%) of the participants admitted to soiling at least once per month: 17% (25) of patients reported soiling daily, 24% (35) weekly, 15% (22) monthly, and 44% (65) less than once a month. There was no difference in the frequency of incontinent episodes between sexes (P = 0.678). However, women who attended the CC reported more frequent leakage, with 27.5% soiling daily compared with 11.0% of those in the UC (P = 0.037), possibly as a result of their bowel condition.

In the CC, significantly more men (79%) disclosed liquid bowel leakage than women (46%), whilst fewer men reported solid (8%) and combined solid/liquid (13%) leakage than women (solid = 18%, combined = 36%; P = 0.008). There was no significant difference between women attending the CC or UC with regard to type of leakage (UC: liquid bowel leakage, 46%; solid bowel leakage, 28%; combined solid/liquid bowel leakage, 25%, P > 0.05). More women attending the CC reported passive leakage (80%) than those attending the UC (62.5%, P = 0.037) or men (48%, P = 0.026). There were no differences between sexes or clinics for fecal urgency or quantity of fecal soiling.

The median (IQR) results of the FIQL scales (range: 1-4; 4 = not affected) for all participants were: lifestyle = 3.24 (2.22-3.80); coping = 2.23 (1.60-3.00); depression = 2.42 (1.95-3.33); and embarrassment = 2.33 (1.67-3.58) (Table 1). Participants who attended the CC had significantly poorer scores than those who attended the UC for lifestyle (P = 0.005), coping (P = 0.003) and embarrassment (P = 0.024) but not depression (P = 0.056).

There were no significant differences in any of the FIQL scales between sexes when compared in total or within the CC (Figure 1). Women who attended the CC had lower scores on all scales compared with those who attended the UC (lifestyle: P = 0.015; coping: P = 0.004; depression: P = 0.062; embarrassment: P = 0.009; Figure 1). The four multiple linear regression analyses [FIQL = clinic + duration + basic FI severity (type × frequency) + error] used to investigate the relationships between QOL of women by clinic, FI type/frequency and duration determined that the poorer QOL results in the CC with regard to clinic and type/frequency remained significant (all P < 0.05), whilst duration of soiling remained significant for the lifestyle, depression and embarrassment scales (all P < 0.05) but not for the coping scale (P = 0.103, Table 2).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Association of FIQL scales with sex and CC and UC. FIQL, Rockwood et al[22]; Box and whisker demonstrates median, IQR, minimum and maximum. P values calculated by unpaired Wilcoxin test (Mann-Whitney); QOL score of 1 represents very low functional status and 4 is not affected by FI; FIQL scales calculated as per letter to the editor, Rockwood[26].

Increasing frequency and quantity of soiling had a significant negative effect across all four QOL scales (frequency, P < 0.001; quantity, P < 0.01). QOL was poorer in participants with greater urgency and passive soiling, whilst those participants who documented both solid and liquid bowel leakage had poorer QOL than those with either alone. Not all scales reached significance (Table 3). When comparing QOL between female participants who attended the CC and UC, the scores for frequency, quantity, type and fecal urgency were significantly poorer across all FIQL scales for female participants in the CC (P < 0.05). Women who reported passive FI in the CC had poorer QOL scores than those in the UC for lifestyle (P = 0.040), coping (P = 0.020), depression (P = 0.181), and embarrassment (P = 0.019).

The model (FIQL = frequency + type + urgency + pad use + error) used to investigate the relationships between the four FIQL scales and the components of severity showed that the significant components that affected QOL were frequency and fecal urgency with regard to lifestyle and coping (all P < 0.05); and frequency and type with regard to depression and embarrassment (all P < 0.05). As a result of the low rate of pad usage (36% wore pads), passive leakage (68% reported) was included in the model, and the quantity of fecal loss was also investigated. This amended model (Table 4) showed: daily, solid, and large losses of stool to be significant factors for lifestyle (P < 0.05); monthly and daily leakage, urgency, pad wearing and large quantities of fecal loss to be significant factors for coping (P < 0.05); frequent and major leakage of both solid and liquid stool were significant factors for depression (P < 0.05); whilst frequent, solid and liquid, and passive stool loss significantly affected embarrassment (P < 0.05).

As many as one in seven adults suffer with fecal incontinence (FI), which can have a devastating effect on the lifestyle of people with frequent, ad hoc or large amounts of fecal seepage. These people often fail to seek treatment due to embarrassment, believing the problem is uniquely theirs, or because they are unaware of the existence of available treatments.

Patients referred to urogynecology and colorectal surgical clinics, for other reasons, have a high incidence of FI. They do not seek assistance whilst attending these clinics from physicians who can recommend or refer for treatment. In this study, the authors demonstrated the impact that FI had on these patients’ quality of life (QOL).

Recent reports have highlighted the impact conservative treatments, such as biofeedback, and more invasive treatments, such as injectable bulking agents, sacral nerve stimulation, artificial bowel sphincter and dynamic graciloplasty, have had on the QOL of patients with FI. This study reports that despite these improvements many people continue to suffer unaware of such advances.

By understanding more than one in five patients attending colorectal and urogynecological clinics have their QOL severely affected by FI, attending physicians can enable access to treatment by directly asking patients about this problem.

FI is the involuntary discharge of solid or liquid stools.

FI severity, including the quantity of stool loss, can have a severe negative impact on QOL. This is an interesting and relevant study that deals with a demanding subject.