From the Department of Pediatrics, Divisions of Gastroenterology and Nutrition (Dr Taubman) and Child Development and Rehabilitation (Dr Blum), University of Pennsylvania, Philadelphia; and the Children's Hospital of Philadelphia (Dr Nemeth).
To evaluate the effect of an intervention targeting parental behavior on stool toileting refusal.
This study population comprised 406 children aged 17 through 19 months from a single suburban private practice. Children were randomly assigned to receive 1 of 2 written toilet-training instructions. Both groups were advised to use a child-oriented approach to toilet training. In addition, those in the intervention group were requested to avoid using negative terms for feces and, before training began, to praise the children when they defecated in the diaper. Follow-up telephone calls were made every 2 to 3 months, and 381 children were followed up until they developed stool toileting refusal or completed daytime toilet training.
The incidence of stool toileting refusal was 23% in the control group and 26% in the intervention group (P>.10). The mean ± SD duration was 5.2 ± 4.9 months in the intervention group vs 7.6 ± 4.9 in the control group (P = .04). Children with stool toileting refusal in the intervention group trained at 40.0 ± 6.4 months vs 43.0 ± 6.5 months in the control group (P = .04).
The intervention did not decrease the incidence of stool toileting refusal but did shorten its duration, leading to earlier completion of toilet training. This should help to ameliorate some of the negative consequences of stool toileting refusal.
DURING THE toilet-training process, as many as 1 in 5 children will go through a period of stool toileting refusal.1 This behavior is associated with several negative consequences. Children with stool toileting refusal train significantly later than their peers, have a higher incidence of stool withholding requiring physician intervention, and are at greater risk for primary encopresis.1,2 An effective intervention reducing the incidence of stool toileting refusal would therefore be desirable.
We have observed that many children hide while defecating in their diaper during the toilet-training process. Some children exhibit this behavior before toilet training has begun. In addition, Schonwald et al3 found that this behavior was common in children with toilet-training problems. This behavior led us to an alternative theory regarding the cause of stool toileting refusal. We hypothesized that children become cognizant of the negative connotation that feces have in our culture through parental verbal and nonverbal signals. This, in turn, causes children to feel shame and embarrassment when defecating, leading them to hide when defecating in the diaper. This also makes children reluctant to defecate in a potty or toilet, because there is invariably at least 1 adult present. To test this theory, we examined the effectiveness of counseling parents not to use negative terms for feces and, before starting toilet training, to praise children when they defecate in the diaper.
All the children were recruited between 17 and 19 months of age from the private practice of one of us (B.T.). This age was chosen because previous research1 suggested parents in this practice would not begin toilet training until several months after this age. Children with global developmental delays or structural abnormalities of the spinal cord, genitourinary, or gastrointestinal tracts were excluded. The practice is in the suburbs of a major metropolitan area and serves predominately middle and upper-middle class families, more than 90% of which are white. Each child's parents gave written informed consent for participation in the study, which was approved by the institutional review board of the Children's Hospital of Philadelphia.
Children were randomly assigned to either the intervention group or the control group. At the children's routine 18-month visit to the pediatrician, parents in both study groups were given handouts counseling them on toilet training. Those in the control group were instructed to use a child-oriented approach to toilet training consistent with published guidelines.4,5 The intervention group received a similar handout and instructions with 2 additions. They were instructed to never use negative terms for feces in the presence of the children. Also, in the months preceding active toilet training for stool, they were asked to praise the children when they defecated in the diaper. All families were enrolled by the same investigator (B.T.), who also gave them their handout and instructions. Any parent questions about toilet training were answered. To minimize possible investigator bias in any further advice about toilet training, all subsequent questions about toilet training were referred to other pediatricians in the practice who were unaware of the treatment group assignment and specific hypotheses of the study.
The Abidin Parenting Stress Index (PSI) was administered at enrollment.6 The PSI measures stress associated with parenting in 2 domains, child and parent. The child domain assesses 4 aspects of the child's temperament and assesses the degree to which the child matches the parent's expectations and is a source of positive reinforcement for the parent. The parent domain assesses 7 sources of stress including parent competence, isolation, attachment, health, role restriction, depression, and spouse relationships.
To evaluate the child's level of language development, the Receptive-Expressive Emergent Language Scale was administered.7 The Receptive-Expressive Emergent Language Scale is a parent report measure designed to assess language development in children from birth to age 36 months. The Receptive-Expressive Emergent Language Scale provides language scores in 2- to 3-month age intervals, which are expressed as developmental quotients (Receptive-Expressive Emergent Language Scale language age/chronological age × 100). Test-retest reliability during a 3-week interval was 0.71.
Follow-up telephone interviews were conducted every 2 to 3 months until the child completed daytime toilet training. Interviewers were blinded as to whether the child was in the intervention group or the control group and did not provide advice about toilet training with one exception. If parents described their child as having significant stool withholding, interviewers instructed the parents to contact their pediatrician. During follow-up interviews, parents were asked where the children urinated and defecated, how many accidents the children had, and whether they were hiding when defecating. In addition, they were questioned about the presence and frequency of constipation or painful defecation. Information about attendance in day care and whether a sibling had been born since the last interview was also obtained.
To assess the integrity of the intervention, parents completed a questionnaire at the 2-year well-child visit. Among other questions about toilet training, the questionnaire asked what terms parents were using to describe feces and what they said to the children when they defecated in the diaper.
Daytime toilet training was considered complete when the child was in underwear whenever awake and had less than 4 urine accidents per week and 2 or fewer episodes of fecal soiling per month. Stool toileting refusal was defined as urinating on the potty or toilet with less than 4 accidents per week but refusing to defecate in the potty or toilet for a period of at least 1 month.
Statistical analyses were performed using Statistical Product and Service Solutions version 9.0 software (SPSS Inc, Chicago, Ill). Means for continuous variables at the baseline assessment and on the follow-up questionnaire were compared in the intervention and control groups using the t test for independent variables. The intervention and control groups differed in sex ratio and parent domain PSI score at baseline. Therefore, when these factors affected a toilet-training outcome, an analysis of covariance procedure was used with the baseline scores as the covariant, group as the main factor, and the toilet-training outcomes as the dependent measure. Dichotomous variables were compared between the 2 groups using the χ2 statistic.
The study population comprised 406 children. Twenty-four children (10 in the intervention group and 14 in the control group) were either lost to follow-up or their parents withdrew from the study. One child from the control group was dropped from the study because of global developmental delay that was not apparent at the time of enrollment. Of the 381 children remaining, 198 were enrolled in the intervention group and 183 in the control group. The characteristics of the 2 study groups are presented in Table 1. Although the intervention and control groups differed at baseline in sex distribution and parent domain PSI score, these variables did not affect the incidence or duration of stool toileting refusal (data not shown).
The incidence of stool toileting refusal and hiding during defecation was not significantly different between the 2 study groups (Table 2). It is possible that those attending day care may have been influenced by the reactions of day-care providers (who were not the target of this intervention) to defecation. Thus, we also evaluated the effect of the intervention in those who never attended day care, but the intervention had no effect on the incidence of stool toileting refusal in this group either (22.8% in the intervention group vs 22.0% in the control group). However, the duration of stool toileting refusal was less in the intervention group (Table 2). Children in the intervention group with stool toileting refusal also completed toilet training earlier than those in the control group. Since both sex and parent domain PSI scores did affect age of toilet-training completion (data not shown), we used an analysis of covariance to investigate the effect of the intervention on age of toilet-training completion in those who developed stool toileting refusal. Those in the intervention group trained earlier whether the analysis was done with or without adjusting for covariates (Table 2, only analysis adjusting for covariates shown). The age at which stool toileting refusal began also was not affected by the intervention.
To assess the integrity of the intervention, 290 parents (159 in the intervention group and 141 in the control group) completed a questionnaire at the 2-year well-child visit. The handouts did affect parents' report of their behaviors, but many in the intervention group did not praise for defecation and some still used negative terms for feces (Table 3).
During the toilet-training process as many as 20% of children develop stool toileting refusal. Of those, one fourth will require intervention.1 Stool toileting refusal is associated with stool withholding, late toilet training, and encopresis.1,2 Several factors have been suggested as causes of stool toileting refusal.8- 11 We know of only 1 study that evaluated a possible cause.11 In a prospective study, Blum at al found that children with stool toileting refusal did not have more behavior problems than other children.
We hypothesized that children may become aware of the negative attitudes of our culture toward feces through verbal and nonverbal signals from the parents. This in turn may cause the child to want to defecate in private, which is difficult for a young child to do when using the potty or toilet. This hypothesis was suggested by our clinical observation that children with stool toileting refusal often hide when defecating in the diaper and the report that up to 70% of children seen at a referral center for children with difficult toilet training hide when defecating.3 Our intervention was designed to eliminate negative signals parents may give to their children regarding feces. We asked parents not to use terms like "stinky diaper" or say the children were dirty or stank when they had a bowel movement. We also asked the parents to verbally praise their children when they defecate in the diaper before active toilet training for stool had begun.
The intervention had no effect on the incidence of stool toileting refusal, but it did have a significant effect on the duration of refusal and, as a consequence, the intervention group trained earlier (Table 2). We believe these differences to be clinically significant. The shorter the duration of stool toileting refusal, the less stress there is on the family and the less likely there is to be a "battle of the bowels."12 It is also probable that the shorter duration of stool toileting refusal, the less likely a child is to develop rectal impaction and primary encopresis.2 This study did not document this, because encopresis occurs in only a small subset of children with stool toileting refusal and we followed families closely recommending intervention if significant stool withholding developed.
It is not clear why the intervention had no effect on the incidence of stool toileting refusal yet decreased the length of time stool toileting refusal was present. Our assessment of the intervention integrity suggested that the degree in which the intervention changed parental behavior was modest, and, given that the assessment was based only on the parents' self-report, actual change may have been less than the parents reported. This modest influence on parental behavior may not be enough to prevent stool toileting refusal but may well have been enough to shorten its duration. Alternatively, stool toileting refusal may result from the interaction of a complex combination of factors of which our intervention addressed only one.
There are some additional limitations of this study. Despite random assignment to intervention and control groups, they did differ in sex ratio and parent domain PSI score. However, these factors did not influence the incidence or duration of stool toileting refusal. This study was done using a population of predominately white middle and upper-middle class families in a single pediatric practice. Factors influencing toilet training may be different in other populations, thereby limiting the ability to generalize the results of this study.13 Furthermore, factors other than those investigated in this study may lead to embarrassment or shame related to defecating. Finally, the use of parental reporting alone in collecting the data may have affected its accuracy.
Although treatments for stool toileting refusal have been described, to our knowledge this is the first prospective study to examine the effectiveness of an intervention designed to decrease the incidence of stool toileting refusal and its consequences.10,11
Corresponding author: Bruce Taubman, MD, Cherry Hill Pediatric Group, 600 W Marlton Pike, Cherry Hill, NJ 08002 (e-mail: firstname.lastname@example.org).
Accepted for publication June 19, 2003.
This is the first prospective study to examine the effectiveness of an intervention designed to decrease the incidence of stool toileting refusal and its consequences. Counseling parents to avoid behavior that may reinforce the negative connotation that feces have in our culture did not decrease the incidence of stool toileting refusal, but it did shorten the time it was present. This in turn resulted in the children toilet training earlier.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Pediatrics editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Register and get free email Table of Contents alerts, saved searches, PowerPoint downloads, CME quizzes, and more
Subscribe for full-text access to content from 1998 forward and a host of useful features
Activate your current subscription (AMA members and current subscribers)
Purchase Online Access to this article for 24 hours
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 13
Customize your page view by dragging & repositioning the boxes below.
and access these and other features:
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.