Abstract
D ysfunctional elimination syndrome (DES) denotes abnormal bowel and bladder voiding with no anatomical or neurological disease present (Colen et al., 2006; Koff, Wagner, & Jayanthi, 1998; Nevéus et al., 2006; Nørgaard, van Gool, Hjälmås, Djurhuus, & Hellström, 1998). It accounts for as many as 30% of referrals to pediatric urologists (Afshar, Mirbagheri, Scott, & MacNeily, 2009). Symptoms of DES include urinary frequency, incontinence, holding maneuvers, urgency, and encopresis (Cooper, 2012; Colen et al., 2006; Nevéus et al., 2006; Nørgaard et al., 1988). DES has also been associated with ineffectual emptying of the bladder, urinary tract infection, and constipation (Desantis, Leonard, Preston, Barrowman, & Guerra, 2011). There is no clear understanding of the etiology of DES. Research has examined associated factors including vesicoureteral reflux (Shaikh et al., 2003), urinary tract infection (Shaikh et al., 2003), toilet training (Schonwald, Sherritt, Stadtler, & Bridgemohan, 2004), behavioral and personality characteristics of the child (Blum, Taubman, & Nemeth, 2003; Joinson, Heron, & von Gontard, 2006), parent perception of child characteristics (Burket et al., 2006), school factors (Cooper et al., 2003), and toileting schedules and bathroom conditions (Lundblad & Hellström, 2005).
Currently, treatment is focused on separate symptoms related to voiding abnormalities. Treatment is typically overseen by urology professionals and includes use of bladder retraining (to frequently and fully empty the bladder), use of laxatives or stool softeners, scheduled voiding, and dietary changes such as high-fiber diets and hydration (Canadian Urological Association, n.d.; Kurzrock & Wagner, n.d.). Other techniques include animated biofeedback and pelvic floor exercises (Kajbafzadeh et al., 2011) and sacral nerve stimulation (Roth, Vandersteen, Hollatz, Inman, & Reinberg, 2008).
Awareness of the need to void the bowel and bladder depends on processing of associated sensory stimuli. Engaging in toileting tasks requires a person to tolerate and respond appropriately to a variety of sensory stimuli. Therefore, when sensory processing is disrupted, toileting may be problematic. Research is needed to explore whether difficulty processing and responding to sensory input, referred to as sensory processing disorder (SPD; Miller, Anzalone, Lane, Cermak, & Osten, 2007), may be related to DES. We proposed two hypotheses to explore this possibility. The primary hypothesis was that children with DES have a higher rate of SPD than the general population. The secondary hypothesis was that a correlation exists between DES and SPD.
Method
Research Design
We used a cross-sectional survey with standardized instruments in a targeted clinical population and a convenience control population. Study procedures were approved by the institutional review boards of The University of Toledo and Mercy St. Vincent Medical Center. Parents or legal guardians of children provided consent to participate and were the sole source of the data collected.
Participants
Two groups of participants were recruited for this study. Recruitment occurred in two medical practice settings: one pediatric urology office and two pediatric offices. The inclusion criteria for both groups were that the child was aged 5–10 yr and that the parent or guardian was at least 18 yr old. Exclusion criteria for both groups included cognitive impairment of the parent or guardian and major medical diagnosis, cognitive impairment, and mental health diagnosis of the child, which were judged by clinical nurse practitioners during the course of the office visit through interview, history, and chart review.
Study Group.
Parents of healthy children ages 5–10 yr seeking treatment of their child’s toileting problems were recruited from a midwestern pediatric urology office if their visit resulted in a subsequent diagnosis of DES for the child.
Control Group.
Parents with healthy children ages 5–10 yr who took their child to their regular pediatrician for a well visit were recruited from two midwestern family practice offices.
Instruments
Vancouver Nonneurogenic Lower Urinary Tract Dysfunction/Dysfunctional Elimination Syndrome Questionnaire.
The Vancouver Nonneurogenic Lower Urinary Tract Dysfunction/Dysfunctional Elimination Syndrome Questionnaire (Vancouver NLUTD/DES; Afshar et al., 2009) includes a total of 14 statements that are responded to on a 5-point Likert scale ranging from 0 (no complaints exist) to 4 (severe symptoms) and can be completed in a few minutes. The questions are worded from the child’s perspective. Afshar et al. (2009) recommended that parents respond for children aged 8 yr and younger. Each statement addresses a single symptom (e.g., urinary frequency, incontinence, constipation). Questionnaire statements relate to frequency of voiding, strain or pain on voiding, and inability to control the bladder and bowels. Each question is weighted equally in computing a total score.
Afshar et al. (2009) reported that test–retest reliability of the Vancouver NLUTD/DES was 84.5% (p = .001). Discriminant function analysis showed 80% accuracy for correctly classifying those with and without DES. Median scores for two groups were statistically significant (p = .001). Cronbach’s α was modest at .445, showing that DES is a heterogeneous clinical syndrome, which is consistent with clinical reports of the condition.
Afshar et al. (2009) used receiver operating characteristic (ROC) curve analysis to determine a cutoff score for diagnosing DES. In ROC curve analysis, scores of participants with and without a known condition are entered into analysis to determine a cut-score that maximizes sensitivity (accurate detection of the condition, or true positive rate) and specificity (diagnosis of only those who do have the condition, or minimization of false-positive rate). The result of an ROC analysis is the area under the curve (AUC) score. AUC scores of 0.9 and higher are deemed excellent. ROC curve analysis for the Vancouver NLUTD/DES resulted in a cut-score of 11, AUC score = 0.903, 95% confidence interval [0.814, 0.948]. A total score of 11 was associated with 80% sensitivity and 91% specificity.
Short Sensory Profile.
The Short Sensory Profile (SSP; McIntosh, Miller, Shyu, & Dunn, 1999) is a tool used to identify sensory processing difficulties in children. It includes 38 statements, which the parent responds to on a 5-point Likert scale (1 = always, 5 = never). The statements and responses assess seven areas of sensory processing: tactile sensitivity, taste and smell sensitivity, movement sensitivity, underresponsive or seeks sensation, auditory filtering, low energy or weak, and visual sensitivity. SSP scores are compared with those of a normative population of 1,037 children without disabilities between ages 3 and 10 yr. According to normal distribution, SSP scores <141 are more than 2 standard deviations from the mean and are categorized as definite difference.
The SSP is a shortened version of the 125-question Sensory Profile (Dunn, 1999). McIntosh et al. (1999) reported that content validity was established through expert consensus and that construct and convergent validity were established through correlation with the School Function Assessment (Coster, Deeney, Haltiwanger, & Haley, 1998). The SSP demonstrated internal reliability for the total test and its sections (Cronbach’s αs = .70–.90).
Procedure
For uniformity in data collection regardless of the children’s ages, the parents or legal guardians were instructed to complete all questionnaires on behalf of their children. Participation was incentivized by a raffle drawing for one of four $50 gift certificates to a child-oriented retail store.
Study Group.
The parent was introduced to the research study at the end of the child’s intake visit by the urology nurse practitioner and asked for consent to participate. If the parent or guardian agreed to participate in the study, he or she was given the forms to complete. Parents remained in the patient room to complete the questionnaires in privacy. Completed surveys were returned before leaving the pediatric urology office.
Control Group.
At the end of a well visit, a nurse practitioner asked parents or guardians whether they would like to participate in survey research. If the parent or guardian agreed, he or she was given a stamped, addressed envelope containing the study materials, including the consent form. Parents who took a packet were asked to provide contact information. After a minimum of 2 wk, a researcher contacted the participants who had not returned completed surveys to encourage them to do so.
Data Analysis
Scoring.
We categorized participants as having DES if their total Vancouver NLUTD/DES score was ≥11. Considering that SPD is not a formal diagnosis (American Academy of Pediatrics [AAP], 2012; Byrne, 2009) and the SSP does not have cut-score for SPD (McIntosh et al., 1999), we operationally defined SPD as having a total SSP score in the definite difference range compared with the mean of the normative population.
Analysis.
The distribution of the SSP scores in the study population was not normally distributed in that it demonstrated kurtosis, so we used nonparametric analyses. For the first hypothesis (that children with DES have a higher rate of SPD than the general population), the frequency of SPD for each group was independently compared with the 5.3% incidence rate in the general population reported by Ahn, Miller, Milberger, and McIntosh (2004) using a statistical test called a nonparametric binomial test. To test the secondary hypothesis (that a correlation exists between DES and SPD), we combined data from both the study and the control group to allow for a range of scores on both assessments. We used a χ2(1) test to assess the correlation of categorization in DES and SPD. This analysis reports the likelihood of having neither, each, or both of the conditions, comparing the likelihoods with chance to determine significance. As a secondary method of testing the second hypothesis, we assessed the correlation between the SSP and the Vancouver NLUTD/DES raw scores using Spearman’s correlation coefficient R s 2, which tests whether the severity of the two conditions are related to each another. For all statistical tests, we determined significance using α = .05.
Results
For the study group, the urology nurse practitioner obtained informed consent from the parents of 21 children; however, 2 did not meet the age criteria. For the control group, 125 research packets were distributed at two general pediatric practices. Fifty-seven were returned for a total return rate of 45.6%. Two control participants were excluded from analysis: 1 participant did not meet the inclusion criteria for age and the other had incomplete data. In total, data from 19 study participants and 55 control participants were entered into analysis.
For the demographic characteristics of the participants, refer to Table 1. The study group was significantly older than the control group by 1.4 yr. The average socioeconomic status of the study group was middle class, and the control group was in the upper middle class. The difference between socioeconomic scores using Hollingshead’s (1975) Four Factor Index was not statistically significant.
Participant Demographics
Note. M = mean; SD = standard deviation; SES = socioeconomic status.
SES was measured with Hollingshead’s (1975) Four Factor Index of Social Status, which takes into account education, occupation, sex, and marital status to determine an estimation of the social status of individuals and, in turn, households. A score is calculated by assessing highest level of education attained, occupation, and whether the family has one or two wage earners. The score determines a generalization that can be made about social status. Scores range from a low of 8 to a high of 66. The higher the score, the more status attributed to that person or household.
One participant did not complete the SES questionnaire.
Differences are significant at p < .05.
Among the control participants, the average age of daytime toilet training was 2.7 yr (standard deviation = 0.7, range = 1–4.20). The average age of nighttime toilet training, of the participants who were trained (n = 50), was 3.3 yr (standard deviation = 1.32, range = 1.0–7.3). Of the 55 control participants, 4 were not yet trained at night, and for 1 the age of nighttime obtainment was not noted. Wald et al. (2009) found that the mean age for initiating toilet training was 27.2 mo, and the mean age of completion was 32.5 mo. Wald et al. and Schum et al. (2002) found that girls trained almost 3 mo earlier than boys. Therefore, the control sample is reflective of typical toilet training patterns.
Higher Rate of Sensory Processing Disorder in Children With Dysfunctional Elimination Syndrome
All study participants had a Vancouver NLUTD/DES score that indicated that they in fact had DES. Of the 19 participants in the study group, 10 (52.6%) also had SPD, as indicated by a score in the definite difference range on the SSP. This proportion of participants in the study group with SPD was significantly higher than the 5.3% rate found in the general population by Ahn et al. (2004; p < .001). In contrast, 4 of 55 (7.3%) participants in the control group had SPD, as indicated by the SSP. The proportion of participants with SPD in the control group was not different from the findings of Ahn et al. (p = .333).
Correlation of Dysfunctional Elimination Syndrome and Sensory Processing Disorder
We found a significant association between the occurrence of DES and SPD, χ2(1) = 20.869, p < .001 (Table 2 ). The likelihood (odds ratio) of having SPD in children with DES was 13/16 (0.81 or 81%); the odds ratio of not having DES for children with SPD was 1/13 (0.08, or 8%); and the odds ratio of having SPD for participants without DES was 1/44 (0.02, or 2%).
Distribution of DES and SPD
Note. DES = dysfunctional elimination syndrome; SPD = sensory processing disorder.
We found a significant negative correlation (Spearman’s ρ = −.493, p < .001; see Figure 1) between SSP and DES raw scores. High Vancouver NLUTD/DES scores (indicative of DES) were associated with lower SSP scores (indicative of SPD), suggesting that the severity of the two conditions is related. The strength of the relationship is defined by the R s 2 value, which was .243. This value approaches the medium effect size of .3 (Cohen, 1992), which suggests that this result is meaningful.
Discussion
According to the results of this study, many children with DES may also have SPD. Given these results, it may be reasonable to include a referral to occupational therapy for assessment and treatment of potential sensory processing problems in addition to the current treatments for DES. Occupational therapists can evaluate functioning in activities of daily living, including toileting. In conjunction with an interdisciplinary team, the therapist and the child’s parents could include the child’s sensory needs as part of their plan for managing the child’s DES. For example, if the child is overresponsive to sensory input, extraneous sensory input in the toileting environment could be minimized, graded and engaging activities could help the child become accustomed to the bathroom environment, and Ayres Sensory Integration© could help normalize the child’s sensory responses (Bellefeuille, Schaaf, & Polo, 2013). Likewise, if the child is underresponsive to sensory input and has low awareness of the need to void, a schedule of timed toileting may be helpful.
Understanding a child’s sensory needs in combination with his or her DES could allow parents to have a greater understanding of their child’s toileting behaviors. The AAP (2011) has reported that toileting accidents are triggers for child abuse because parents struggle to set reasonable expectations for their children and interpret accidents as willful disobedience. Creating insight into the factors that contribute to toileting accidents may reduce the risk for abuse associated with toileting accidents (AAP, 2011).
Limitations and Future Research
Several limitations in the interpretation of these findings are acknowledged. First is the use of targeted and convenience samples, as well as the study group sample size. Also, as a result of intermittent increased patient volume at the study site, not every child who met the inclusion criteria was offered the opportunity to participate in this research. These factors limit the reliability and generalization of the findings and should motivate larger studies. Second, the method of data collection also introduces some limitations. The information provided through parent report could introduce potential bias. Finally, it is possible that parents of children in the study group demonstrated a negative response bias in completing the questionnaires. Perhaps the stress that their child’s bowel and bladder difficulties place on daily life caused them to view their child as being more difficult, sensitive, or stubborn, resulting in a higher rate of endorsing symptoms of SPD.
Given the limitations of this study, replication studies are warranted. In doing so, researchers may want to use the full version of the Sensory Profile (Dunn, 1999) because factor analysis would allow exploration of the relationship between DES and subtypes of SPD. Future studies may address the relationship of DES and SPD from the opposite perspective by investigating whether children who are in treatment for SPD also have a higher rate of DES. Research is needed to clarify whether DES and SPD have a causal relationship or shared causation for both conditions. For clinical practice, it will be important to investigate sensory-based treatment of DES. Following from the case study by Bellefeuille et al. (2013), further studies could determine whether treating both DES and SPD simultaneously could result in improved outcomes.
Implications for Occupational Therapy Practice
The results of this study have the following implications for occupational therapy practice:
Children with DES are likely to also have SPD. This study found that a majority of children with DES seen in one clinic also presented with SPD, which could not be accounted for by a regional difference because the rate of SPD in a control sample did not vary from that in a previous report on the general population (Ahn et al., 2004).
Referral to occupational therapy may be appropriate to address this aspect of care in an interdisciplinary team (AAP, 2012; Byrne, 2009). Understanding a child’s unique sensory processing can help practitioners and parents have a greater understanding of the child’s toileting behaviors.
Because elimination disorders have considerable impact on children and their families, expanding the understanding of how to develop a successful treatment plan that is client centered is important.
Footnotes
Acknowledgments
We thank the staff at St. Vincent Mercy Medical Center, Wood County Medical Associates, and University Pediatrics at the University of Toledo Medical Center, in particular Melissa Kurtz, Marjean Beck, Janell Arps, and Thomas Russell.