Journal of Pediatric Psychology Advance Access originally published online on February 7, 2008
Journal of Pediatric Psychology 2008 33(7):739-750; doi:10.1093/jpepsy/jsn008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
This article appears in the following Journal of Pediatric Psychology issue: Series on Methodology Article [View the issue table of contents]
Early Childhood Risk Factors Associated with Daytime Wetting and Soiling in School-age Children
1Department of Community Based Medicine, University of Bristol, 2Department of Social Medicine, University of Bristol, 3Department of Child and Adolescent Psychiatry, Saarland University Hospital, 4Sheffield Children's; NHS Foundation Trust, and 5Department of Community Based Medicine, University of Bristol
All correspondence concerning this article should be addressed to Dr Carol Joinson, ALSPAC-Avon Longitudinal Study of Parents and Children, 24 Tyndall Avenue, Bristol, BS8 1TQ, UK. E-mail: carol.joinson{at}bristol.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionSupplementary DataAcknowledgmentsReferences |
|---|
Objective To examine the impact of a range of early childhood factors on the risk for daytime wetting and soiling. Methods This is a longitudinal study based on a UK population of over 10,000 children from age 4 to 9 years. Parents completed questionnaires on child development (at 18 months); child temperament (24 months); maternal depression/anxiety (21 months), and parenting behaviors (24 months). The analysis examined whether these risk factors distinguish between children with normal development of daytime bladder and bowel control and those with delayed acquisition of daytime continence; persistent daytime wetting/soiling, and relapse in wetting/soiling. Results Delayed development, difficult temperament, and maternal depression/anxiety were associated with an increase in the odds of experiencing problems with bladder and bowel control. Conclusions The current findings provide evidence that risk factors in early childhood are associated with a subsequent increase in the odds of children experiencing daytime wetting and soiling at school age.
Key words: ALSPAC; bladder and bowel control; child development; daytime wetting; developmental trajectory; longitudinal study; maternal depression; soiling.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionSupplementary DataAcknowledgmentsReferences |
|---|
The attainment of daytime bladder and bowel control is an essential milestone of early childhood development. Previous studies describing general trends in the development of daytime bladder and bowel control find that wetting and/or soiling accidents are not uncommon in preschool children, but the majority are reliably dry and clean during the day between the ages of 3 and 5 years (Brazelton, 1962
; Largo, Molinari, von Siebenthal, Wolfensberger, 1999; Schum et al., 2002). However, these studies do not adequately capture the considerable variability between individuals in their pathways to daytime continence. For example, some children experience delays in achieving daytime bladder and bowel control and continue to have wetting and/or soiling accidents after the age of 4 years; some children may experience a relapse after having been previously dry and clean, and others will continue to suffer from persistent wetting and/or soiling problems well into their school years. Daytime incontinence, due to its public nature, is likely to be accompanied by embarrassment and ridicule from peers and has been shown to be associated with emotional and psychological problems in school-age children in the community (Joinson, Heron, Butler, von Gontard, 2006; Joinson, Heron, von Gontard, 2006). Despite the severe consequences associated with daytime wetting and soiling, there is currently little evidence-based information from epidemiological studies on risk factors that distinguish between children who attain daytime bladder and bowel control as expected and those who continue to suffer from wetting and soiling at school age.
It has been suggested that delayed neurologic development may be associated with problems in attaining continence, since control over the bladder and bowels not only requires complex coordination between the nerves and muscles, but also the development of the necessary motor and communication skills needed to achieve independent toileting (Brazelton, 1962; Brazelton et al., 1999). There is evidence that delayed development of language and motor skills is more common in children with wetting problems compared to controls, but the findings were based on small samples comprising children with mixed daytime and nighttime wetting (Jarvelin, 1989; Jarvelin et al., 1991). Developmental delays have also been reported in children with soiling (Foreman & Thambirajah, 1996), but the findings are limited because they were based on a case-note study of only 63 boys and there was no control group. Another study reported no association between delayed motor and language milestones and attainment of daytime bladder and bowel control, but the sample was relatively small (Largo et al., 1999). The association between developmental milestones and the attainment of daytime bladder and bowel control clearly requires further investigation in a representative sample of children from the community.
Child temperament is another factor that may impact on the development of toileting. It is possible that difficult child temperament traits may result in conflicts between the parent and child during the toilet training process and lead to the development of toileting problems. More difficult temperament traits were found in children who were reported by parents to be difficult to toilet train, including children who were not toilet trained at 3
to 4 years of age (after 6 months of trying), and those between 4 and 7 years of age who refused to use the toilet or who had not yet completed daytime bladder or bowel training (Schonwald, Sherritt, Stadtler, & Bridgemohan, 2004). Children with stool toileting refusal (urinating in the toilet but refusing to use the toilet for bowel movements) have also been reported to have more difficult temperament (Blum, Taubman, & Osborne, 1997), while another study reported no association between difficult child temperament and later age at completion of toilet training (Blum, Taubman, & Nemeth, 2004). However, these studies are limited because they are all based on relatively small clinic samples. Moreover, the studies reporting more difficult temperament traits may have been biased, because one study comprised children referred to a clinic for children with wetting, soiling, and failure to toilet train (Schonwald et al., 2004), and the other comprised children of parents who were actively involved in a study of toilet training (Blum et al., 1997).
There are well-documented effects of maternal depression and anxiety on child development (Downey & Coyne, 1990; Ross & McLean, 2006; Rutter, 1989), but no studies have specifically investigated whether these effects extend to the development of toileting problems. Having a mother with depression or anxiety is a potentially stressful experience for a child because it is associated with increased hostility and irritability toward the child, lack of sensitivity, or responsiveness to the child's needs and an increased likelihood of engaging in harsh disciplining behaviors (McLearn, Minkovitz, Strobino, Marks, & Hou, 2006; Nicol-Harper, Harver, & Stein, 2007; Sohr-Preston & Scaramella, 2006). Previous studies of children who have problems attaining nocturnal bladder control have found evidence that early stress is an important etiological factor for bedwetting (Douglas 1973; Jarvelin, Moilanen, Vikevainen-Tervonen, & Huttunen, 1990; MacKeith, 1968). Psychosocial stressors occurring during the sensitive stage of toilet training may make children susceptible to developing problems with bladder control (Fergusson, Horwood, & Shannon, 1990; Jarvelin et al., 1990). However, this research has not been extended to investigate whether early psychosocial stressors associated with exposure to maternal depression have an impact on the development of daytime bladder and bowel control.
It has also been suggested that factors relating to parenting may be associated with the development of bladder and bowel problems in children and that certain parenting styles or behaviors may interfere with the toilet training process. Parents who engage in "ongoing harmful approaches" and those who are "mishandling toilet training problems" for example, by "punishing the child for non-compliance or forcibly holding the child on the toilet" (Schmitt, 2004, p. 72), may put their child at increased risk of developing problems with toileting. Only two clinic-based studies have examined specific parenting factors in relation to the development of children's toileting problems, but they failed to find strong evidence for an association with "dysfunctional parenting styles" including "laxness," "verbosity," and "over-reactivity" (Schonwald et al., 2004) or parenting stresses including perceived parental competence, isolation, and spouse relationships (Blum et al., 2004). However, these findings do not rule out the possibility that other types of parenting behaviors may be associated with the development of toileting problems in children.
In order to increase understanding of the heterogeneity in the development of daytime bladder and bowel control, a useful starting point is to investigate the variability in individual pathways to continence. This is important because it may help to elucidate different risk factors relating to characteristics of individuals and their environments that account for qualitative differences in their development of bladder and bowel control. The purpose of the current study is to examine whether early childhood risk factors discriminate between children who develop daytime bladder and bowel control as expected and those who continue to suffer from daytime wetting and soiling into their school years. It is hypothesized that the following risk factors will be associated with increased odds of developing wetting and soiling problems at school age: developmental delay; difficult child temperament; maternal depression and anxiety, and parenting behaviors.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionSupplementary DataAcknowledgmentsReferences |
|---|
Participants
The Avon Longitudinal Study of Parents and Children (ALSPAC) is a longitudinal, population-based birth cohort study that recruited 14,541 pregnant women resident in Avon, UK with expected dates of delivery April 1, 1991 to December 31, 1992. There were 14,062 live born children. The study protocol and the representative nature of the ALSPAC sample have been described previously (Golding, Pembrey, & Jones, 2001
). Further details are on the ALSPAC website: http://www.alspac.bris.ac.uk. Ethical approval for the study was obtained from the ALSPAC Law and Ethics Committee and the Local Research Ethics Committees. Data were collected using self-completion questionnaires sent to the mothers during pregnancy and approximately annually since the birth of the child. Direct assessments were carried out at annual research clinics from the age of 7 years.
Measures
Questions relating to the child's daytime wetting and soiling were asked regularly between 15 months and 9 years. The current study used pairs of responses from questionnaires administered at 4, 5, 6, 7, and 9 years and questions were phrased in such a way that an occurrence of daytime wetting or soiling was considered unusual rather than normal. At 4 and 5 years, the carer was asked how often the child wet his/herself (or dirtied his/her pants) during the day with the following options: (a) never; (b) less than once a week; (c) about once a week; (d) more than once a week; and (e) every day. At 6, 7, and 9 years, options (d) and (e) were replaced by (d) 2–5 times a week, (e) nearly every day, and (f) more than once a day. For the purpose of these analyses, responses were collapsed into simple "yes/no" binary variables.
Development at 18 months was assessed using a questionnaire developed by ALSPAC including items from the Denver Developmental Screening Test (Frankenburg, Dodds, Archer, Shapiro, & Bresnick, 1992) comprising four domains (gross motor, 
= .689; fine motor, = .679; communication, = .752; and social skills, = .624). The Toddler Temperament Scale (Fullard, McDevitt, & Carey, 1984) was used to assess nine temperamental traits at 24 months (activity, = .593; adaptability, = .637; approach, = .848; distractibility, = .712; intensity, = .570; mood, = .672; persistence, = .711; rhythmicity, = .672; and threshold, = .517). Maternal anxiety and depression were assessed when the study child was 21 months using the Crown-Crisp Experiential Index (previously known as the Middlesex Hospital Questionnaire), comprising depression ( = .706), anxiety ( = .771), and somatic symptoms ( = .524) scales (CCEI; Crown & Crisp, 1979; Crown, Duncan, & Howell, 1970) together with the 10 questions from the Edinburgh Postnatal Depression Scale ( = .867; EPDS; Cox, Holden, & Sagovsky, 1987). Parenting behavior was assessed when the study children were 24-months old using a set of 10 questions derived for the ALSPAC study asking parents how frequently they engaged in certain behaviors relating to nurturing ( = .560) and discipline practices ( = .668). Further information on these measures can be found in the Supplementary Appendix on the Journal of Pediatric Psychology Website.
Statistical Modelling
Recent years have seen the development of powerful new statistical methodologies for modelling longitudinal data. These methods, referred to as "group-based trajectory modelling" (Muthén & Muthén, 1998–2006; Nagin & Tremblay, 2005), are becoming increasingly popular in epidemiological research and have been used to elucidate trajectories to nighttime bladder control (Croudace, Jarvelin, Wadsworth, & Jones, 2003). One such method, Longitudinal Latent Class Analysis, has been used recently to empirically derive trajectory groups for daytime wetting and soiling (Heron, Joinson, Croudace, & von Gontard, 2008) and it is these results that the current manuscript builds on.
LLCA is a person-centered approach that models the set of responses obtained for each participant, e.g., child A might have responses "Yes, No, No, No, and No" for the questions on daytime wetting at 4, 5, 6, 7, and 9 years respectively, whereas child B might have responses "Yes, Yes, Yes, Yes, and No". Such variability in these patterns of responses is accounted for by a latent factor that groups together children exhibiting similar patterns of development. The result is a number of different latent classes, here referred to as "trajectory groups". Each group can be seen to exhibit its own trajectory through time by plotting age against the class-specific prevalences of daytime wetting or soiling. Since some children are more easily assigned to a particular trajectory group than others, the output from a latent class model is a set of posterior probabilities describing the likelihood of a child belonging to each group. For instance, child A would be likely to have a high probability of being assigned to a group characterized by normal or slightly delayed development, whilst child B would have much lower probability of being a member of one of these groups, favouring a group characterized by persistent daytime wetting.
Models were carried out separately on repeated binary measures (as described earlier) of daytime wetting and then daytime soiling. Results showed that the variability in daytime wetting and soiling could be adequately described by dividing the children into four trajectory groups for each condition. For daytime wetting, a group showing normative development of daytime bladder control comprised 86.2% of the population and was characterized by a low (<10%) prevalence of daytime wetting at 54 months and a negligible prevalence thereafter. The other daytime wetting groups were as follows: delayed attainment of daytime bladder control (6.9%; steadily decreasing prevalence of daytime wetting from 80% at 54 months through 40% at 78 months down to <10% prevalence by 9 years); persistent daytime wetting (3.7%; prevalence of daytime wetting >80% until 91 months and a reduction to 60% at 9 years), and relapse in daytime wetting following a period of previously being dry (3.2%; prevalence of daytime wetting of 40% at 54 months, <10% prevalence at 65 months rising to 60% at 78 months and declining slowly thereafter). For bowel control, the trajectories defining the four groups were very similar in shape and interpretation: normative development of daytime bowel control (89%); delayed attainment of daytime bowel control (4.1%); persistent daytime soiling (2.7%); and relapse in daytime soiling following a period of previously being clean (4.1%).
For the purpose of the model described in this article, the two variables representing these trajectory groups were used as outcome variables in an examination of the risk factors (child development; child temperament; maternal depression and anxiety; and parenting behaviors) that might distinguish between normal and atypical development of daytime bladder and bowel control. This was achieved using a set of multinomial models in which the odds ratios were estimated free of constraint (e.g., no proportional odds assumption), and derived in relation to the normative trajectories of daytime bladder and bowel control, which were used as the reference groups. The data set was reshaped so that there were four data points for each child, indicating the posterior probability assignment to each latent class, and then these posterior probabilities were used as a weighting (e.g., if a particular child was assigned a probability of .7 for class 1 and .1 for the remaining classes, then the class 1 observation would be given seven times the weight in the analysis). Bootstrapping was employed during preliminary work to sample from each child's distribution of posterior probabilities and this demonstrated that standard errors obtained using the weighting approach were accurate (details available from first author on request).
Regression estimates were adjusted for gender and a number of indicators of social adversity experienced during the first 2 years of the study child's life (comprising early parenthood, housing adequacy, maternal education, financial difficulties, family size, partner support, and the presence of an emotional support network). Further adjustment was made for children with an IQ < 70 (obtained from assessment at a research clinic the children attended at 8 years using the Weschler Intelligence Scale for Children Third Edition– WISC-III; Wechsler, Golombok, & Rust, 1992) due to the possibility that low IQ may be associated with both delayed attainment of continence as well as the early childhood risk factors. Analyses were repeated excluding children with an IQ < 70 to examine whether the findings were being driven by a disproportionate number of children with low IQ among the atypical trajectory groups. Initial trajectory models were fitted using Mplus version 4.2 (Muthén & Muthén, 1998–2006) and multinomial models were fitted with Stata version 9.2 using "mlogit" command with the "iweight" option.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionSupplementary DataAcknowledgmentsReferences |
|---|
The starting cohort consisted of 13,793 (singletons/twins) who were still alive at 1 year of age. For the latent class analysis of daytime wetting and daytime soiling, data were available on at least one measurement occasion for 10,819 and 10,821 children, respectively. Consequently, these figures were the starting sample for the multinomial models in the current study. For each of the individual models, the sample was restricted to those who had a response to the risk factor under consideration. There were only minor differences in the distribution of classes among the cases used in the final models compared with those lost through missing covariates. For instance, among the cases used in the adjusted models for the indices of child development, the relapse group made up 3.2%, persistent 3.7%, delayed 6.9%, and normative 86.2%, whereas within the cases lost through attrition, the distribution was 3.0%, 3.9%, 6.7%, and 86.3% of the respective classes.
Daytime Wetting
The results for the daytime wetting models are shown in Table I. The table shows both unadjusted and adjusted odds ratios and 95% confidence intervals for the association of the risk factors (development; temperament; maternal depression and anxiety; and parenting) and the different trajectories of daytime bladder control (relapse, persistent, and delayed). Odds ratios were derived in relation to the normative trajectory of daytime bladder control (used as the reference group). For continuous risk factors (temperament/development), regression estimates are for 1 SD increase in the covariate. The p-values shown are for a global test of differences across the four outcome groups, i.e., all the three odds ratios shown are equal to unity. There was strong evidence for an effect in all domains of developmental delay (p < .001 throughout). A mild strengthening of effects following adjustment was found to be due to the different gender ratio between trajectory groups. Effects were largest for the persistent wetting group with an increase of 1 SD in indices of developmental delay being associated with an increase in odds of between 33% (communication) and 53% (total development) in the adjusted model. More moderate effects were observed for delayed bladder control with the increase in odds between 17% for fine motor development and 30% for total development. Effects for the relapse group were weak with the lower bound of the confidence intervals indicating that the effect could be either negligible or even in the opposite direction.
|
A similar pattern is observed for the temperament traits, although effect sizes are smaller. In particular, there was strong evidence for differences in odds for the domains of "adaptability," "mood," "persistence," and "threshold" (p < .001) and to a lesser extent "intensity" (p = .013). Again, there was little evidence for an increase in odds of suffering a relapse in daytime wetting being related to an increase in scores in any of these five, or any of the other temperament domains. Adjustment for the confounding variables (gender, social adversity, IQ < 70) had little effect on the magnitude of the estimates.
Effects appear much stronger for maternal depression and anxiety reflecting the binary nature of these risk factors. Evidence is strong for all four of the maternal psychopathology factors considered. For EPDS and CCEI anxiety, the magnitude of effect estimates are similar across the three outcome levels shown, particularly for "CCEI anxiety" with odds ratios for the daytime wetting trajectories of 1.44 for relapse, 1.49 for persistent, and 1.54 for delayed bladder control. In contrast, high scores for "CCEI depression" and "CCEI somatic" show a stronger association with the persistent daytime wetting group. Finally, there is little evidence for parenting behaviors relating to nurturing or disciplining being associated with an increase in the odds of following an atypical trajectory for the development of daytime bladder control.
Daytime Soiling
The results for the daytime soiling models are shown in Table II. Once again there is strong evidence for an effect in all domains of developmental delay. Odds ratios were highest for the persistent group with the combined (total) development domain having the strongest effect, as with daytime wetting. It is noticeable in the results for developmental delay that the odds ratios for a relapse in soiling are larger than those found for daytime wetting, and they are comparable in magnitude to those for delayed bowel control. Lower confidence limits rule out the possibility of a negative effect (reduction in odds) for the relapsing group. Adjustment for confounders led to more marked attenuation in the odds ratios, reflecting the fact that boys strongly out-number girls within all three atypical soiling classes (Heron et al., 2008). Again there is strong evidence for differences in the temperament traits of "adaptability," "mood," "persistence," and "threshold" and it is apparent that an increase of 1 SD in indices of difficult temperament is associated with an increase in odds of belonging to the relapse class as well as the delayed and persistent classes.
|
Effects were again much stronger for maternal depression and anxiety, with the persistent daytime soiling group showing the strongest associations. Effects for the relapse category are stronger, whilst there is little evidence for an increase in the odds of belonging to the delayed class. The analysis involving parenting behaviors again yielded little evidence of differences across classes.
Children with an IQ < 70
Among the 10,824 children who had a measure of trajectory membership, 114 (1.05%) had an IQ < 70. Within this low IQ subgroup, no children had Down's syndrome compared with seven children in the group with an IQ of 70 or more. Eighty-seven in this subgroup were assessed using the Development and Well Being Assessment at age 7 years (Goodman, Ford, Richards, Gatward, & Meltzer, 2000). Of these children, 12 (14.1%) met DSM-IV criteria for ADHD; 7 (8.2%) met criteria for oppositional-conduct disorder and 2 (2.3%) met criteria for pervasive development disorder. Special Educational Needs (SEN) status was established in 2002/2003 for 108 of the subgroup as part of the Pupil Level Annual School Census (PLASC) and 26 (24.1%) were reported to require no special provision; 31 (28.7%) were classed as "school action," 20 (18.5%) "school action plus," and 31 (28.7%) had a statement of SEN. More information on SEN and PLASC can be found at www.teachernet.gov.uk/sen/ and www.schoolsweb.gov.uk/locate/management/tar/plasc/. Finally, three children (2.6%) in the subgroup were identified from The National Health Service notes of ALSPAC children with a suspected developmental disorder and/or statement of SEN as meeting criteria for an autism diagnosis.
Having an IQ < 70 predicted trajectory membership in the following way: daytime wetting—relapse = 1.34 [0.51, 3.49]; persistent = 2.51 [1.28, 4.94]; delayed =1.51 [0.80, 2.85], global p = .042; daytime soiling—relapse = 2.05 [1.01, 4.15]; persistent = 3.55 [1.80, 4.15]; delayed = 1.37 [0.58, 3.20], global p = .001. Due to the possibility that any of the observed associations were being driven by the low IQ subgroup, the analysis was repeated excluding these 114 cases. As expected, among the children with IQ < 70, the relative size of each trajectory group was markedly different to the children with normal IQ—over 20% showed atypical development of daytime bowel control, with a similar proportion for daytime bladder control—however, the small size of the excluded group meant that the class distributions across the sample as a whole remained unaffected by this exclusion. For the majority of the analyses shown in Tables I and II, the effect of removing these children was negligible—typically a change of 0.01 or 0.02 in the unadjusted regression estimates. The effect was slightly more marked for the maternal depression/anxiety measures but the attenuation was still only in the region of 10%. None of the conclusions were altered by the exclusion of children with an IQ < 70 and a similar picture emerged with the fully adjusted models (figures available on request).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionSupplementary DataAcknowledgmentsReferences |
|---|
This is the first longitudinal population-based study of early childhood risk factors associated with developmental trajectories of daytime wetting and soiling in school age children. Using previously derived developmental trajectories of daytime bladder and bowel control in children aged 4–9 years (Heron et al., 2008
), the study examined the association of the trajectories with child and parent variables measured when the child was between 18- and 24-months old.
The current findings provide evidence that children with delayed development in motor, communication, and social skills and those with difficult temperament traits (e.g., less adaptable; negative mood) are more likely to experience daytime wetting and soiling into their school years. Delayed development and difficult temperament may disrupt the toilet training process and result in persistent toileting problems throughout childhood. Parents may benefit from being encouraged to consider their child's developmental level and temperament in planning toilet training strategies, because different strategies will suit different children. For example, children who are less adaptable to new situations may resist parental pressure for sudden change in their routine, e.g., from using nappies (diapers in US) to using a toilet or potty, and such struggles may be associated with the development of toileting problems. It is also possible that difficult temperament is an early indicator of persistent externalizing or internalizing problems in children with bladder and bowel control problems. Previous studies have reported that early temperament is related to the development of child psychopathology (Muris & Ollendick, 2005; Rettew & McKee, 2005) and daytime wetting and soiling have previously been shown to be associated with externalizing and internalizing problems in school age children in the ALSPAC cohort (Joinson, Heron, Butler, et al., 2006; Joinson, Heron, & von Gontard, 2006).
There are well-established long-term effects of maternal psychopathology on child development (Rutter, 1989), but this is the first study that we are aware of to find evidence for an impact of maternal depression and anxiety on the development of daytime bladder and bowel control. In particular, the odds of exhibiting persistent daytime wetting or soiling were higher among children of mothers who reported symptoms of depression and/or anxiety when their children were around 21 months of age. The timing of depressive/anxiety symptoms coincides with the age when many children begin to acquire bladder and bowel control and when many parents will have initiated toilet training. Toilet training is a challenging and stressful developmental phase (Schmitt, 1987) and is likely to place increased demands on mothers who are emotionally vulnerable. Mothers’ psychological distress may result in stress in the child if the mother shows lack of sensitivity to their child's needs and negative or inconsistent responses to their child's efforts during toilet training. Stress occurring during the sensitive stage of toilet training may interrupt the acquisition of bladder and bowel control and make children susceptible to developing wetting and soiling problems.
The current study also examined different parenting behaviors (frequency of nurturing or disciplining), but did not find any strong evidence that differences in these behaviors relate to the development of daytime wetting and soiling problems. However, these findings do not rule out the possibility that parenting behaviors not yet examined, or those specifically relating to toilet training practices may be associated with the development of toileting problems.
The trajectory group characterized by relapse in daytime wetting generally showed the weakest associations with early development and temperament. More proximal factors such as urinary tract infections (Meadow, 1990) and recent emotional stress (e.g., starting a new school; birth of a new sibling) (Sureshkumar, Craig, Roy, & Knight, 2000) are perhaps more potent risk factors for relapse in daytime wetting in school-age children. There was some evidence for an association between relapse in soiling and maternal reports of delayed early development and difficult temperament, suggesting that these early risk factors may make children more vulnerable to subsequent relapses in soiling during the school years. Maternal psychopathology was associated with an increased risk of relapse in daytime wetting and soiling in school age children. It is possible that mothers who reported earlier depression and/or anxiety (when their child was 21 months) were more likely to have these symptoms during the period of assessment of daytime wetting and soiling in the current study (when children were 4- to 9-years old) and these more proximal symptoms could be a risk factor for relapse.
A subgroup of children with IQ < 70 was included in the study due to the well-known relationship between low IQ and daytime incontinence (von Wendt, Simila, Niskanen, & Jarvelin, 1990). It was felt that excluding these cases might affect the generalizability of the findings, particularly for the groups with persistent wetting or soiling throughout early childhood. To investigate the possibility that the findings were being driven by children with low IQ, the analysis was repeated excluding these cases, but this did not alter the main conclusions. This suggests that among children with normal intellect, there are associations between daytime incontinence and early childhood risk factors, and that the effects are not purely due to the overrepresentation of children with low IQ in the persistent, delayed, and relapse trajectory groups.
Limitations
A limitation of the current study is that the developmental trajectories of daytime bladder and bowel control were based on measures of wetting and soiling from 4 to 9 years. Ideally, the trajectories should have started earlier to allow the examination of more subtle differences in attainment of bladder and bowel control that were still within the normal range (i.e., differences between children falling into the normative trajectory groups). However, this was not possible because comparable measures of daytime wetting and soiling were not available prior to age 4 years.
At the other end of the time period examined in the current study (9 years of age), a proportion of children had still not attained daytime bladder and bowel control. Only a small proportion of children continue to suffer from persistent wetting and/or soiling problems into adolescence and beyond, and it is possible that many of the group denoted here as "persistent" would be re-defined as "delayed" or "severely delayed" if it was possible to extend the age range of the trajectories. Availability of a further time point would perhaps result in a shrinking of the persistent group, with a corresponding strengthening of the model estimates for the risk factors considered.
In deriving the binary measures of daytime wetting and soiling, we have discarded information on frequency. The purpose of this study was to show the different developmental trajectories of bowel and bladder control in a nonclinical population and not to focus on children whose daytime wetting or soiling meets DSM-IV criteria. Previous studies on the ALSPAC population have shown that it is important to consider children whose daytime wetting and soiling are not severe enough to meet established DSM-IV criteria because these less severe groups also have a number of adverse outcomes (Joinson, Heron, Butler, et al., 2006; Joinson, Heron, & von Gontard, 2006).
Another possible limitation of this study is that it relied on maternal reports for both wetting and soiling and the risk factors relating to the child and mother. Relying on parental reports of children's toileting behaviour would be potentially more of a problem if the study had examined frequency of wetting/soiling, due to the possibility that this could be under/overestimated by parents. The use of binary measures of daytime wetting and soiling should have minimized this limitation and it is unlikely that a child's wetting or soiling would go entirely unnoticed by parents. It is also possible that parents of children who are wetting or soiling may overestimate problems relating to their child, such as difficult temperament. However, the retrospective design of this study means that parental reports of child temperament predate the reports of wetting and soiling problems. Mothers suffering from depression are more likely to view their child negatively (Reck et al., 2004) and this is potentially more of a problem, because maternal depression (assessed when the child was 21 months) may have negatively influenced maternal reports of child temperament (assessed at 24 months).
The effect sizes found in the present study are small probably because the wetting and soiling groups were not restricted to the most severe cases that are likely to present to clinics, but also comprised children with daytime wetting and soiling that did not meet DSM-IV criteria. However, the purpose of the current study was to examine risk factors for daytime wetting and soiling across the whole range of wetting/soiling problems in a sample drawn from the general population. It is notable that a previous study of children with daytime wetting found evidence for neurological impairments in children whose wetting did not meet DSM-IV criteria (Jarvelin et al., 1991). It is also notable that the findings relating to temperament in the current study were in the same specific domains as those reported in a previous study (adaptability, negative mood, persistence; Schonwald et al., 2004). In addition, to these findings, the current study has shown that the effects of delayed development and difficult temperament can differ between the "atypical" (i.e., persistent, delayed, and relapse) wetting/soiling pathways.
The methods used here have enabled us to dramatically simplify the complex set of responses relating to the occurrence of daytime wetting and soiling observed within this cohort. We have represented the changing patterns relating to the development of daytime bladder and bowel control of over 10,000 children as four groups with some level of communality in their progression towards continence and have then examined differences in the etiology of these groups. This modeling approach to describing population heterogeneity is simple to understand and offers a good account of these data. Although such models can often describe data very well in terms of groups, that does not imply that these groups actually exist, merely that this representation of the data is useful and appealing, especially from the point of view of prognosis and treatment. There is always scientific value in replication of the subgroups in other data sets, where additional scientific scrutiny can be gained through external validation against other criteria, such as clinical diagnoses of daytime wetting and soiling.
Clinical Implications
The study found strong evidence for an increased risk of atypical development of daytime bladder and bowel control in children whose mothers reported suffering from depression and/or anxiety at 21 months. These findings have important implications for the pediatricians’ role in detecting maternal psychological distress due to its impact on child development. Although mothers often recognize the potential impact of depression on their child's health and well being (Kahn et al., 1999), maternal depression often goes unrecognized by pediatricians due to lack of training in its detection and diagnosis (Olson et al., 2002). The contribution of the child's temperament/behavior to parental depression is another factor that should be considered by clinicians. Mothers of children with behavior problems are at risk of increased levels of depressive symptoms (Gartstein & Sheeber, 2004). Pediatricians should be wary of particular constellations of child and parent risk factors, including child behavior problems and mothers’ evaluation of parental competence, because these may exacerbate maternal depressive symptomatology. The results of the current study underscore the importance of regular developmental checks on children of depressed mothers so that anticipatory guidance can be offered to avoid potential parent–child conflicts that may arise during stressful phases such as toilet training.
The current findings provide an increased understanding of the development of bladder and bowel control in children in the community, as well as providing evidence-based information on risk factors associated with daytime wetting and soiling. Increased knowledge of the risk factors associated with atypical trajectories of daytime continence is an essential tool for healthcare workers involved in the treatment of children with daytime wetting and soiling. An important minority of children suffer from wetting and soiling into the school years, and if these problems remain unresolved or untreated they can become socially and psychologically debilitating (Issenman, Filmer, & Gorski, 1999).
Important areas to consider for future research include investigating whether the risk factors identified in the current study anticipate children with increased vulnerability to problems with bladder and bowel control and whether interventions targeting these factors result in a decreased risk of developing toileting problems. Future studies should also extend the age range covered in the current study to investigate risk factors associated with wetting and soiling problems that persist into adolescence and adulthood, and examine other child and parent risk factors, such as prenatal influences, diet, constipation, family history, and toilet training practices.
|
Supplementary Data |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionSupplementary DataAcknowledgmentsReferences |
|---|
Supplementary data are available at JPEPSY Online.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionSupplementary DataAcknowledgmentsReferences |
|---|
We are extremely grateful to all the families who took part in this study, the midwives for their help in recruiting them, and the whole ALSPAC team, which includes interviewers, computer and laboratory technicians, clerical workers, research scientists, volunteers, managers, receptionists, and nurses. The UK Medical Research Council, the Wellcome Trust, and the University of Bristol provide core support for ALSPAC. This publication is the work of the authors who also serve as guarantors for the contents of this article. This research was specifically funded by a grant from the Big Lottery Fund and has been carried out in collaboration with the charity ERIC (Education and Resources for Improving Childhood Continence).
Conflicts of interest: None declared.
Received September 4, 2007; revision received January 18, 2008; accepted January 18, 2008
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionSupplementary DataAcknowledgmentsReferences |
|---|
Blum NJ, Taubman B, Nemeth N. Why is toilet training occurring at older ages? A study of factors associated with later training. Journal of Pediatrics (2004) 145(1):107–111.[CrossRef][Web of Science][Medline]
Blum J, Taubman B, Osborne ML. Behavioural characteristics of children with stool toileting refusal. Pediatrics (1997) 99(1):5–53.
Brazelton TB. A child-oriented approach to toilet training. Pediatrics (1962) 29:121–128.
Brazelton TB, Christophersen ER, Frauman AC, Gorski PA, Poole JM, Stadtler AC, et al. Instruction, timeliness, and medical influences affecting toilet training. Pediatrics (1999) 103(6 Pt 2):1353–1358.
Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. British Journal of Psychiatry (1987) 150:782–786.
Croudace TJ, Jarvelin MR, Wadsworth ME, Jones PB. Developmental typology of trajectories to nighttime bladder control: Epidemiologic application of longitudinal latent class analysis. American Journal of Epidemiology (2003) 157(9):834–842.
Crown S, Crisp AH. Manual of the crown-crisp experiential index (1979) London: Hodder and Stoughton.
Crown S, Duncan KP, Howell RW. Further evaluation of the Middlesex hospital questionnaire (M.H.Q.). British Journal of Psychiatry (1970) 116:33–37.
Douglas JWB. Early disturbing events and later enuresis. In: Bladder control and enuresis—Kolvin I, MacKeith RC, Meadows SR, eds. (1973) London: William Heinemann Medical Books. 109–117.
Downey G, Coyne JC. Children of depressed parents: An integrative review. Psychological Bulletin (1990) 108(1):50–76.[CrossRef][Web of Science][Medline]
Fergusson DM, Horwood LJ, Shannon FT. Secondary enuresis in a birth cohort of New Zealand children. Paediatric and Perinatal Epidemiology (1990) 4:53–63.[Medline]
Foreman DM, Thambirajah MS. Conduct disorder, enuresis and specific developmental delays in two types of encopresis: A case-note study of 63 boys. European Child and Adolescent Psychiatry (1996) 5(1):33–37.[CrossRef][Web of Science][Medline]
Frankenburg WK, Dodds J, Archer P, Shapiro H, Bresnick B. The Denver II: A major revision and restandardization of the Denver Developmental Screening Test. Pediatrics (1992) 89(1):91–97.
Fullard W, McDevitt SC, Carey WB. Assessing temperament in one- to three-year-old children. Journal of Pediatric Psychology (1984) 9(2):205–217.
Gartstein MA, Sheeber L. Child behavior problems and maternal symptoms of depression: A mediational model. Journal of Child and Adolescent Psychiatric Nursing (2004) 17(4):141–150.[CrossRef]
Golding J, Pembrey M, Jones R. ALSPAC—the Avon Longitudinal Study of Parents and Children. I. Study methodology. Paediatric and Perinatal Epidemiology (2001) 15(1):74–87.[CrossRef][Web of Science][Medline]
Goodman R, Ford T, Richards H, Gatward R, Meltzer H. The development and well-being assessment: Description and initial validation of an integrated assessment of child and adolescent psychopathology. Journal of Child Psychology and Psychiatry (2000) 41:645–655.[CrossRef][Web of Science][Medline]
Heron J, Joinson C, Croudace T, von Gontard A. Trajectories of daytime wetting and soiling in a UK population of 4 to 9 year olds. Journal of Urology (2008) (in press).
Issenman RM, Filmer RB, Gorski PA. A review of bowel and bladder control development in children: How gastrointestinal and urologic conditions relate to problems in toilet training. Pediatrics (1999) 103(6 Pt 2):1346–1352.
Järvelin MR. Developmental history and neurological findings in enuretic children. Developmental Medicine and Child Neurology (1989) 31:728–736.[Web of Science][Medline]
Jarvelin MR, Moilanen I, Vikevainen-Tervonen L, Huttunen N. Life changes and protective capacities in enuretic and non-enuretic children. Journal of Child Psychology and Psychiatry (1990) 31(5):763–774.[CrossRef][Medline]
Jarvelin MR, Moilanen I, Kangas P, Moring K, Vikevainen-Tervonen L, Huttunen N, et al. Aetiological and precipitating factors for childhood enuresis. Acta Paediatrica Scandinavica (1991) 80:361–369.[Web of Science][Medline]
Joinson C, Heron J, Butler U, von Gontard A. Psychological differences between children with and withtout soiling problems. Pediatrics (2006) 117(5):1575–1584.
Joinson C, Heron J, von Gontard A. Psychological problems in children with daytime wetting. Pediatrics (2006) 118(5):1985–1993.
Kahn RS, Wise PH, Finkelstein JA, Berstein HH, Lowe JA, Homer CJ. The scope of unmet maternal health needs in pediatric settings. Pediatrics (1999) 103:576–581.
Largo RH, Molinari L, von Siebenthal K, Wolfensberger U. Development of bladder and bowel control: Significance of prematurity, perinatal risk factors, psychomotor development and gender. European Journal of Pediatrics (1999) 158(2):115–122.[CrossRef][Medline]
McLearn KT, Minkovitz CS, Strobino DM, Marks E, Hou W. The timing of maternal depressive symptoms and mothers’ parenting practices with young children: Implications for pediatric practice. Pediatrics (2006) 118(1):e174–e182.
MacKeith RC. A frequent factor in origins of primary nocturnal enuresis: Anxiety in the third year of life. Developmental Medicine and Child Neurology (1968) 10:465–470.[Medline]
Meadow SR. Day wetting. Pediatric Nephrology (1990) 4(2):178–184.[CrossRef][Web of Science][Medline]
Muthén LK, Muthén BO. Mplus User's Guide, Fourth Edition (1998–2006) Los Angeles, CA: Muthén & Muthén.
Muris P, Ollendick TH. The role of temperament in the etiology of child psychopathology. Clinical Child and Family Psychology Review (2005) 8(4):271–289.[CrossRef][Medline]
Nagin DS, Tremblay R. What has been learned from group-based trajectory modeling? Examples from physical aggression and other problem behaviors. Annals of the American Academy of Political and Social Science (2005) 602:82–117.
Nicol-Harper R, Harvey AG, Stein A. Interactions between mothers and infants: Impact of maternal anxiety. Infant Behav Dev (2007) 30(1):161–167. [Epub September 12, 2006].[Medline]
Olson AL, Kemper KJ, Kelleher KJ, Hammond CS, Zuckerman BS, Dietrich AJ. Primary care pediatricians’ roles and perceived responsibilities in the identification and management of maternal depression. Pediatrics (2002) 110:1169–1176.
Reck C, Hunt A, Fuchs T, Weiss R, Noon A, Moehler E, et al. Interactive regulation of affect in postpartum depressed mothers and their infants: An overview. Psychopathology (2004) 37:272–280.[CrossRef][Web of Science][Medline]
Rettew DC, McKee L. Temperament and its role in developmental psychopathology. Harvard Review of Psychiatry (2005) 13(1):14–27.[CrossRef][Medline]
Ross LE, McLean LM. Anxiety disorders during pregnancy and the postpartum period: A systematic review. Journal of Clinical Psychiatry (2006) 67(8):1285–1298.
Rutter M. Psychiatric disorder in parents as a risk factor for children. In: Prevention of mental disorder, alcohol and other drug use in children and adolescents—Schaffer D, Phillips I, Enger NB, eds. (1989) Rockville, Maryland: Office for Substance Abuse, USDHHS.
Schmitt BD. Seven deadly sins of childhood: Advising parents about difficult developmental phases. Child Abuse & Neglect (1987) 11(3):421–432.[CrossRef][Web of Science][Medline]
Schmitt BD. Toilet training problems: Underachievers, refusers, and stool holders. Contemporary Pediatrics (2004) 21(4):71–82.
Schonwald A, Sherritt L, Stadtler A, Bridgemohan C. Factors associated with difficult toilet training. Pediatrics (2004) 113(6):1753–1757.
Schum TR, Kolb TM, McAuliffe TL, Simms MD, Underhill RL, Lewis M. Sequential acquisition of toilet-training skills: A descriptive study of gender and age differences in normal children. Pediatrics (2002) 109(3):E48.[CrossRef][Medline]
Sohr-Preston SL, Scaramella LV. Implications of timing of maternal depressive symptoms for early cognitive and language development. Clinical Child and Family Psychological Review (2006) 9(1):65–83.[CrossRef]
Sureshkumar P, Craig JC, Roy LP, Knight JF. Daytime urinary incontinence in primary school children: A population-based survey. Journal of Pediatrics (2000) 137(6):814–818.[CrossRef][Web of Science][Medline]
von Wendt L, Simila S, Niskanen P, Jarvelin MR. Development of bowel and bladder control in the mentally retarded. Developmental Medicine and Child Neurology (1990) 32(6):515–518.[Web of Science][Medline]
Wechsler D, Golombok S, Rust J. WISC-III UK. In: Wechsler intelligence scale for children – Third Edition UK Manual (1992) Sidcup, UK: The Psychological Corporation.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||