Fecal incontinence—the involuntary leakage of liquid or solid stool—has been estimated to affect approximately 9% of women during the first 3 months postpartum.1 There is ongoing controversy about the contribution of method of birth to postpartum fecal incontinence. A systematic review of studies examining the effect of mode of birth on bowel symptoms showed that instrumental vaginal birth and spontaneous vaginal birth were associated with increased risk of fecal incontinence when compared with cesarean delivery.2 In contrast, a recent Cochrane review found no difference in risk of fecal incontinence after a cesarean delivery compared with vaginal birth.3 These reviews are limited by the paucity and quality of included studies. Methodologic weaknesses of this literature include the use of nonstandardized outcome measures for assessing anal and fecal incontinence, confusion regarding the precise definition of fecal incontinence symptoms, lack of information regarding symptoms that predate the index pregnancy and birth, and limited capacity to distinguish early symptoms (3 months postpartum or less) from symptoms that persist beyond the first 3 months after birth.
The Maternal Health Study is a multicenter, prospective, nulliparous pregnancy cohort study designed to investigate postpartum health problems including fecal incontinence. Women were recruited during early pregnancy and followed-up at 3-month intervals until their first child was 12 months old. The aims of the current article are to investigate whether postpartum fecal incontinence is less common in women who had a cesarean delivery and more common in women who had an operative vaginal birth compared with women who had a spontaneous vaginal birth, and to investigate whether postpartum fecal incontinence is more common in women who report fear of an intimate partner during pregnancy or abuse by an intimate partner in the first 12 months postpartum. Aim I was prespecified in the study protocol.4 Aim II extends analyses previously reported for this cohort, showing an association between fear of an intimate partner and fecal incontinence in early pregnancy.5
MATERIALS AND METHODS
Nulliparous women were recruited to the study via six public maternity hospitals in Melbourne, Australia, between April 2003 and December 2005. Eligibility criteria were age 18 years or older, capacity to complete questionnaires and telephone interviews in English, and an estimated gestation of 24 weeks or less at enrollment (according to date of last menstrual period or ultrasound examination). Participants were asked to agree to complete two short computer-assisted telephone interviews at 30–32 weeks of gestation and 9 months postpartum and written questionnaires at 3, 6, and 12 months postpartum. Data on pregnancy, labor, and birth were also abstracted from study hospital medical records for participants who gave written consent.
Recruitment was facilitated by participating hospitals with staff sending invitation packages, which included a copy of the baseline questionnaire, to women soon after booking to give birth in a study hospital. A single mailed reminder postcard was sent to nonresponders 2 weeks after the initial mailing. We were not able to follow-up nonresponders by telephone because of Australian privacy legislation. Women were invited to return the questionnaire, consent papers, and contact information in a reply paid envelope. Given the intensive follow-up over an extended period asked of participants, coupled with the potentially sensitive nature of questions on bowel symptoms and sexual health, we anticipated an overall response fraction of between 20% and 40%, consistent with other recent longitudinal cohort studies.6,7 Although acknowledging any level of nonresponse may lead to bias, a response fraction in this range was deemed an acceptable trade-off provided that high rates of follow-up were achieved (80% or more). Empirical evidence to support the assumption that exposure–outcome associations remain reliable in studies with relatively low response fractions is provided by two recent studies.8,9
Sample size calculations based on prevalence estimates from previous studies10,11 and allowing for 10% loss to follow-up showed that a sample of 1,500 women would be sufficient to test the hypothesis that women having an operative vaginal birth have a twofold greater relative risk of development of fecal incontinence compared with women having a spontaneous vaginal birth (8% compared with 2.6%, α=0.05 [two-sided], β=0.80).5
The type, frequency, and severity of fecal incontinence were assessed using standardized questions based on instruments previously validated in Australian, Scandinavian, and United Kingdom populations.12 – 14 The baseline questionnaire included questions about symptoms since the start of pregnancy and symptoms in the 12 months before the pregnancy. The computer-assisted telephone interview at 30–32 weeks of gestation asked about symptoms in the previous month. Postpartum questionnaires and the telephone interview at 9 months asked about symptoms in the previous 3 months. Women were defined as having fecal incontinence if they reported any involuntary leakage of liquid or solid bowel movements after standardized terminology for bowel symptoms endorsed by the International Continence Society.15 For period prevalence, women were classified as incontinent if they reported fecal incontinence at any follow-up in a defined period (eg, first 12 months) and continent if they reported no incontinence in all completed follow-up questionnaires and computer-assisted telephone interviews within the defined period. Women completing less than two postpartum follow-up visits were classified as missing rather than continent. Fecal incontinence severity was defined as low, moderate, or severe based on the frequency and amount of loss drawing on the urinary incontinence scoring system devised by Sandvik et al13,14 and modified for fecal symptoms. For the purpose of analysis, we distinguished between postpartum fecal incontinence in the first 3 months postpartum and fecal incontinence symptoms present between 4 and 12 months postpartum. Data were also collected on flatal incontinence (report of involuntary loss of flatus) and fecal urgency (sudden compelling desire to defecate that is difficult to defer) as defined by the International Continence Society.12 Women reporting an urgent need to open their bowels that made them rush to the toilet at least once per week were defined as having fecal urgency incontinence. Details regarding question wording and the method of scoring are published elsewhere.5
Intimate-partner violence was assessed using the short 18-item version of the Composite Abuse Scale, which asks about physical and emotional abuse.16,17 The Composite Abuse Scale was included in the 12-month follow-up questionnaire, and women were asked to indicate how frequently different types of abusive behavior had happened to them since the birth of their newborn (ie, over the course of a 12-month period) on a scale that included the following possible responses: never, only once, several times, once per month, once per week, and daily. At each follow-up, women were also asked to report on whether they were currently afraid of a partner or former partner or whether they had “ever” been afraid of an intimate partner. An Australian validation of the Composite Abuse Scale showed that these questions have good sensitivity and specificity for identifying women who have experienced physical abuse or severe combined physical and sexual abuse, but only moderate sensitivity for identifying emotional abuse alone.16
Data on pregnancy, labor, and birth events were obtained from hospital medical records and questionnaires. Medical record data were used to determine the timing of cesarean delivery, length of labor, onset of second stage, use of analgesia, degree of genital tract trauma, and time of birth. Onset of second-stage labor was defined as full dilatation of the cervix when recorded, or based on commencement of pushing, involuntary urge to push, or presenting part on view with contractions, when information regarding dilatation of the cervix was not available. Prolonged second-stage labor was defined as more than 2 hours from onset of second-stage labor to the birth with no regional analgesia (epidural, spinal, or combined) or more than 3 hours from the onset of second-stage labor to the birth with regional analgesia.18,19
Questionnaire data were used in preference to medical records for previous terminations or miscarriages, maternal weight before pregnancy, and maternal position in active second-stage labor (upright or not upright). Prepregnancy body mass index (BMI, calculated as weight (kg)/[height (m)]2) was classified as underweight if BMI was less than 18.5, as normal if 18.5–24.9, as overweight if 25–29.9, and as obese if 30 or more.
To assess the representativeness of the sample, we compared data on the social and obstetric characteristics of participants with routinely collected Victorian perinatal data for nulliparous women giving birth as public patients in Victoria during the recruitment period. Data were analyzed using STATA 11.1. Period prevalence of fecal incontinence in the 12 months before the index pregnancy, during early to late pregnancy, and during the first 12 months postpartum were based on the proportion of women reporting incontinence divided by the total number of women completing each stage of follow-up. Logistic regression was used to report on associations between fecal incontinence at 4 to 12 months postpartum, method of birth, and other obstetric and maternal factors. In a secondary analysis, logistic regression was used to examine associations between fecal incontinence, intimate-partner violence (Composite Abuse Scale), fear of an intimate partner during or after pregnancy, and fear of an intimate partner before the index pregnancy. Multivariable logistic regression was used to obtain a more precise estimate of the association between method of birth (as the exposure of main interest), intimate-partner violence, and fecal incontinence at 4 to 12 months postpartum.20 Other factors included to account for potential confounding were maternal age, prepregnancy BMI, and the hospital in which the birth occurred. Data are presented as crude or adjusted odds ratios (ORs) with 95% confidence intervals (CIs).
The study was approved by the institutional research committees and ethics committees of the participating hospitals: Royal Women's Hospital, Melbourne (2002/23); Southern Health, Melbourne (2002–099B); and Angliss Hospital, Melbourne; and by the research ethics committees of La Trobe University (2002/38) and the Royal Children's Hospital, Melbourne (27056A).
The baseline sample comprised 1,507 eligible women. It was not possible to calculate a precise response fraction because hospitals were responsible for managing the mailing of invitation packages. This meant we were not able to determine exact numbers of invitation packages that were incorrectly addressed or sent in duplicate. A conservative estimate based on 80% to 90% of invitation packages being sent to eligible women gives a final response fraction of 28% to 31%, but this may be an underestimate. The mean gestation of participants at the time of completing the baseline questionnaire was 15.0 weeks (standard deviation 3.1, range 6–24 weeks).
As shown in Table 1, participants were representative of Victorian women giving birth in the study period in relation to method of birth and neonatal birth weight. However, lower responses were received from young women (18–24 years, 15.5% compared with 29.9%) and women born overseas with non–English-speaking backgrounds (16.2% compared with 21.0%).
Follow-up response fractions of 90% or higher were achieved at all follow-up points. In late pregnancy (30–32 weeks of gestation), data were provided by 1,454 of the 1,483 women who had not given birth by 32 weeks of gestation (98%). At 3, 6, 9, and 12 months postpartum, response fractions were 95% (1,431 of 1,507), 93% (1,400 of 1,507), 92% (1,387 of 1,507), and 90% (1,357 of 1,507), respectively. The sample for this article comprised the 1,296 women who completed the 3-month follow-up and who had medical record data abstracted. Multiple births (n=13) were excluded, leaving a sample of 1,283 women ranging from 19 to 50 years of age at the time of their newborn's birth (mean age 31.0 years, standard deviation 4.9 years).
Just more than 7% of women (94 of 1,282) reported that they had fecal incontinence in the 12 months before pregnancy and 6.5% (83 of 1,282) reported it during pregnancy, one-third of whom (28 of 83) also reported fecal incontinence before pregnancy. The proportions of women reporting anal, fecal, and urge incontinence at 3, 6, 9, and 12 months postpartum and severity of symptoms are shown in Table 2. A majority of women who reported fecal incontinence in questionnaires reported symptoms of low severity, whereas those disclosing fecal incontinence in the computer-assisted telephone interview were more likely to describe symptoms of moderate severity.
Seventeen percent of women reported fecal incontinence at some point in the first 12 months postpartum, with most of these women reporting episodic symptoms (ie, symptoms at one or two time points only). Eight percent of women reported fecal incontinence in the first 3 months postpartum; 42.6% of these women (43 of 101) reported fecal incontinence in a subsequent follow-up. Of the 12.8% of women who reported one or more episodes of fecal incontinence between 4 to 12 months postpartum, the majority (114 of 157; 72.6%) had not reported fecal incontinence in the first 3 months postpartum.
In early pregnancy, 2.8% (36 of 1,273) of women reported that they were currently afraid of an intimate partner, and 15.6% (198 of 1,273) said they had ever been afraid of an intimate partner. In total, 6.1% (77 of 1,269) of women reported that they were afraid of their partner during their pregnancy (combining data collected at study enrollment in early pregnancy with data collected retrospectively at 6 and 9 months postpartum).
Women's responses to the Composite Abuse Scale indicated that 16% of women (190 of 1,187) experienced abuse by an intimate partner in the first 12 months postpartum. The majority (54.2%; 103 of 190) reported emotional abuse alone, 32.6% (62 of 190) reported emotional and physical abuse, and 13.2% (25 of 190) reported physical abuse alone.
Women who had a vaginal birth assisted with forceps or vacuum extraction were significantly more likely than women who had a spontaneous vaginal birth to report fecal incontinence in the first 3 months postpartum (OR 1.93 and 95% CI 1.09–3.41, and OR 2.08 and 95% CI 1.20–3.61, respectively). Women who had a cesarean delivery had significantly reduced odds of reporting fecal incontinence in the first 3 months postpartum (OR 0.51, 95% CI 0.28–0.93) compared with women who had a spontaneous vaginal birth.
Table 3 reports data showing the association between fecal incontinence at 4 to 12 months postpartum, method of birth, and other maternal and obstetric covariates. Women who had a cesarean delivery had slightly increased odds of reporting fecal incontinence at 4 to 12 months postpartum compared with women who had a spontaneous vaginal birth (OR 1.14, 95% CI 0.77–1.69). Operative vaginal birth was also associated with slightly increased odds of fecal incontinence compared with spontaneous vaginal birth. However, these differences were not statistically significant. Factors associated with significantly increased odds of reporting fecal incontinence at 4 to 12 months postpartum were: fecal incontinence reported before pregnancy, fecal incontinence reported in pregnancy, being underweight (BMI less than 18.5), and older maternal age (35 years or older; Table 3). Among women who had a third-degree or fourth-degree tear, there was a twofold increase in odds of fecal incontinence at 4 to 12 months postpartum compared with women who had a vaginal birth without a third-degree or fourth-degree tear (22.9% [11 of 48] compared with 12.1% [101 of 833], OR 2.15, CI 1.1–4.4). In addition to method of birth, obstetric covariates not associated with fecal incontinence 4 to 12 months postpartum included: use of epidural or spinal analgesia, hours of pushing in second stage, prolonged second-stage labor, and maternal upright position in labor. Results for episiotomy (in most cases medio-lateral) showed a borderline increase in odds.
The unadjusted associations between postpartum fecal incontinence and variables assessing intimate-partner violence are shown in Table 4. Fecal incontinence at 4 to 12 months postpartum was significantly more common among women who had experienced emotional or physical violence in the first 12 months postpartum as measured on the Composite Abuse Scale. Women who were afraid of a partner during pregnancy and those afraid of a partner in the year after giving birth were also significantly more likely to report fecal incontinence at 4 to 12 months postpartum.
To obtain more precise estimates of the association between method of birth and fecal incontinence at 4 to 12 months postpartum, we developed a multivariable logistic regression model with fecal incontinence as the outcome variable (Table 5). Method of birth (cesarean, operative vaginal birth, spontaneous vaginal birth) and intimate-partner violence were the exposures of main interest. Fecal incontinence reported in pregnancy was included because of the significant association with fecal incontinence noted in univariable analyses. Intimate-partner violence was associated with significantly increased odds of reporting fecal incontinence 4 to 12 months postpartum in the adjusted model (adjusted OR 1.62, 95% CI 1.05–2.50).
It has been hypothesized that cesarean delivery may prevent postpartum fecal incontinence by avoiding direct vaginal, sphincter, and distal pelvic nerve trauma.3,21,22 The majority of studies investigating this hypothesis focus on symptoms in the first 3 months postpartum.2,3 The major contribution of the current study is to provide a more complete and detailed picture of the natural history of fecal incontinence and other bowel symptoms over the course of the first postpartum year. We deliberately chose to focus on fecal incontinence beyond 3 months postpartum because this is a gap in existing literature. Our findings show that women who had a cesarean delivery for their first birth did not have a lesser likelihood of reporting fecal incontinence at 4 to 12 months postpartum. Nor was there any greater likelihood of fecal incontinence associated with operative vaginal birth compared with spontaneous vaginal birth. Women who had a third-degree or fourth-degree tear had increased odds of fecal incontinence at 4 to 12 months postpartum but accounted for only 6.9% (11 of 159) of women with fecal incontinence beyond 3 months postpartum. No statistically significant associations were found for other labor and birth factors.
Factors that did significantly increase the likelihood of fecal incontinence at 4 to 12 months postpartum were fecal incontinence in pregnancy, being underweight, and experiencing intimate-partner violence in the first 12 months postpartum. Women reporting fecal incontinence in pregnancy were three to four times more likely to report postpartum fecal incontinence after adjusting for method of birth and other covariates. Most previous studies assessing the potential contribution of method of birth to postpartum fecal incontinence have not taken predelivery fecal incontinence into account. In the current study, 17% (27 of 159) of women who reported fecal incontinence at 4 to 12 months postpartum also reported symptoms during pregnancy. Women who reported fecal incontinence in pregnancy were more likely to give birth by elective cesarean delivery. It is possible that pre-existing symptoms may influence some women and clinicians to opt for an elective cesarean delivery for a first birth.
This study examines the relationship between intimate-partner violence and postpartum fecal incontinence. One in five women with fecal incontinence at 4 to 12 months postpartum had experienced intimate-partner violence during the year after giving birth. Abuse by an intimate partner is a recognized risk factor for physical and psychological health problems during and after pregnancy23 – 25 but has not been previously considered as a potential risk factor for postpartum fecal incontinence. Comorbid anxiety and a higher occurrence of irritable bowel syndrome among women experiencing intimate-partner violence may partially explain this association.26,27 It is also likely that previous and current sexual abuse may play a role in postpartum fecal incontinence, although no studies specifically examine this pathway.28,29 Women reporting fecal incontinence in pregnancy were more likely to have an elective cesarean delivery, but not more likely to have a cesarean delivery once labor commenced compared with those who were nonsymptomatic in pregnancy (data not shown). Women who reported intimate-partner violence during the year after the index birth were also more likely to have given birth by cesarean delivery. The sample size for this study, although relatively large compared with the majority of previous studies, is not sufficient to permit further detailed exploration of causal pathways taking these potential mediating factors into account. Major strengths of this study are the recruitment of a nulliparous sample in early pregnancy, frequent follow-up to 12 months postpartum, and use of standardized measures for assessing fecal incontinence and intimate-partner violence. In addition, data on labor events were obtained from both hospital medical records and women's own accounts. Although larger than most studies examining associations between obstetric events and postpartum fecal incontinence, the sample size provided limited power for several exposure comparisons of interest and is inadequate for detailed examination of complex causal pathways.
Comparison of participant characteristics with routinely collected data for births at the six study hospitals confirmed that the sample is representative with regard to method of birth, neonatal birth weight, and other obstetric characteristics, but it is under-representative of younger women and women of non–English-speaking backgrounds.5 Over-representation of older women may have resulted in slightly higher prevalence of fecal incontinence in our sample but is likely to have been offset by exclusion of women attending late for antenatal care known to be at higher risk for intimate-partner violence. There was no evidence of differential recruitment of women with pre-existing bowel disease compared with community prevalence studies, and our estimate of the prevalence of fecal incontinence in the first 3 months postpartum accords with the findings of other large cohort studies.1
We deliberately chose to assess associations with symptoms at 4 to 12 months postpartum to exclude women whose experience of postpartum fecal incontinence was confined to the first few weeks after giving birth. Although our findings confirmed results of other studies showing associations between use of forceps and fecal incontinence at 6 to 12 weeks postpartum,2 evidence from the current study suggests that method of birth does not play a major role in fecal incontinence at 4 to 12 months postpartum.
Understanding causal pathways for postpartum fecal incontinence requires attention to the interplay of pregnancy and birth events and upstream factors such as intimate-partner violence. In this nulliparous cohort, method of birth was not a major determinant of fecal incontinence status beyond 3 months postpartum. This information needs to be considered when discussing the evidence regarding risks and benefits of cesarean delivery. Clinicians also need to be aware of the potential contribution of intimate-partner violence to postpartum fecal incontinence.
1. MacArthur C, Glazener CM, Wilson PD, Herbison GP, Gee H, Lang GD, et al.. Obstetric practice and faecal incontinence three months after delivery. BJOG 2001;108:678–83.
2. Pretlove SJ, Thompson PJ, Toozs-Hobson PM, Radley S, Khan KS. Does the mode of delivery predispose women to anal incontinence in the first year postpartum? A comparative systematic review. BJOG 2008;115:421–34.
3. Nelson RL, Furner SE, Westercamp M, Farquhar C. Cesarean delivery for the prevention of anal incontinence. The Cochrane Database of Systematic Reviews 2010, Issue 2. Art. No.: CD006756. DOI: 10.1002/14651858.CD006756.pub2.
4. Brown SJ, Lumley JM, McDonald EA, Krastev AH. Maternal Health Study: a prospective cohort study of nulliparous women recruited in early pregnancy. BMC Pregnancy Childbirth 2006;6:12.
5. Brown SJ, McDonald EA, Krastev AH. Fear of an intimate partner and women's health in early pregnancy: fecal incontinencendings from the Maternal Health Study. Birth 2008;35:293–302.
6. 45 Up Study Collaborators. Cohort profecal incontinencele: the 45 and up study. Int J Epidemiol 2008;37:941–7.
7. Magnus P, Irgens LM, Haug K, Nystad W, Skjaerven R, Stoltenberg C, et al.. Cohort profecal incontinencele: the Norwegian Mother and Child Cohort Study (MoBa). Int J Epidemiol 2006;35:1146–50.
8. Martikainen P, Laaksonen M, Piha K, Lallukka T. Does survey non-response bias the association between occupational social class and health? Scand J Public Health 2007;35:212–5.
9. Mealing NM, Banks E, Jorm LR, Steel DG, Clements MS, Rogers KD. Investigation of relative risk estimates from studies of the same population with contrasting response rates and designs. BMC Med Res Methodol 2010;10:26.
10. Brown S, Lumley J. Maternal health after childbirth: results of an Australian population based survey. Br J Obstet Gynaecol 1998;105:156–61.
11. MacArthur C, Bick DE, Keighley MR. Faecal incontinence after childbirth. Br J Obstet Gynaecol 1997;104:46–50.
12. Gunthorpe W, Brown W, Redman S. The development and evaluation of an incontinence screening questionnaire for female primary care. Neurourol Urodyn 2000;19:595–607.
13. Sandvik H, Seim A, Vanvik A, Hunskaar S. A severity index for epidemiological surveys of female urinary incontinence: comparison with 48-hour pad-weighing tests. Neurourol Urodyn 2000;19:137–45.
14. Sandvik H, Hunskaar S, Seim A, Hermstad R, Vanvik A, Bratt H. Validation of a severity index in female urinary incontinence and its implementation in an epidemiological survey. J Epidemiol Community Health 1993;47:497–9.
15. Haylen BT, de Ridder D, Freeman RM, Swift SE, Berghmans B, Lee J, et al.. An International Urogynecological Association (IUGA)/International Continence Society (ICS) joint report on the terminology for female pelvic floor dysfunction. Neurourol Urodyn 2010;29:4–20.
16. Hegarty K, Fracgp, Bush R, Sheehan M. The composite abuse scale: further development and assessment of reliability and validity of a multidimensional partner abuse measure in clinical settings. Violence Vict 2005;20:529–47.
17. Hegarty K, Sheehan M, Schonfeld C. A multidimensional defecal incontinencenition of partner abuse: Development and preliminary validation of the Composite Abuse Scale. J Fam Violence 1999;14:399–415.
18. American College of Obstetricians and Gynecologists. Operative vaginal delivery: use of forceps and vacuum extractors for operative vaginal delivery. ACOG Practice Bulletin No. 17. Washington, DC: ACOG; 2000.
20. Hosmer DW, Lemeshow S. Applied logistic regression. 2nd ed. New York: John Wiley & Sons, Inc; 2000.
21. Sand PK. Should women be offered elective cesarean section in the hope of preserving pelvic floor function? Int Urogynecol J Pelvic Floor Dysfunct 2005;16:255–6.
22. Nygaard I. Should women be offered elective cesarean section in the hope of preserving pelvic floor function? Int Urogynecol J Pelvic Floor Dysfunct 2005;16:253–4.
23. Silverman JG, Decker MR, Reed E, Raj A. Intimate partner violence victimization prior to and during pregnancy among women residing in 26 U.S. states: associations with maternal and neonatal health. Am J Obstet Gynecol 2006;195:140–8.
24. Bacchus L, Mezey G, Bewley S. Domestic violence: prevalence in pregnant women and associations with physical and psychological health. Eur J Obstet Gynecol Reprod Biol 2004;113:6–11.
25. McFarlane J, Parker B, Soeken K. Abuse during pregnancy: associations with maternal health and infant birth weight. Nurs Res 1996;45:37–42.
26. Donnelly VS, O'Herlihy C, Campbell DM, O'Connell PR. Postpartum fecal incontinence is more common in women with irritable bowel syndrome. Dis Colon Rectum 1998;41:586–9.
27. Wang J, Varma MG, Creasman JM, Subak LL, Brown JS, Thom DH, et al.. Pelvic floor disorders and quality of life in women with self-reported irritable bowel syndrome. Aliment Pharmacol Ther 2010;31:424–31.
28. Leserman J, Drossman DA, Li Z, Toomey TC, Nachman G, Glogau L. Sexual and physical abuse history in gastroenterology practice: how types of abuse impact health status. Psychosom Med 1996;58:4–15.
29. Beck JJ, Elzevier HW, Pelger RC, Putter H, Voorham-van der Zalm PJ. Multiple pelvic floor complaints are correlated with sexual abuse history. J Sex Med 2009;6:193–8.