Routinized and compulsive-like behaviors (RCB) are common in typically developing children and in children and adults with Down syndrome, but what functions do they serve? Parents completed questionnaires for RCB, behavior problems, and adaptive behaviors. Children who had Down syndrome had significantly higher levels of RCB than did the typically developing children at all MAs; RCBs were positively associated with adaptive behaviors for younger MA and CA groups, but not older MA children and all adults with Down syndrome. For children with Down syndrome and MAs over 5 years and all adults, RCBs were associated with behavior problems. We concluded that RCBs support developmental progress for all children with MAs less than 5 years, but may have different functions for older individuals.
A pervasive issue in intellectual disability is whether behaviors are developmentally appropriate or pathological. Researchers have reported misdiagnosis of psychiatric illness in individuals with Down syndrome who exhibited self-talk, had imaginary friends, or fantasized about TV stars (e.g., Glenn & Cunningham, 2000; Hurley, 1996). However, one can argue that these behaviors are typical developmental phenomena and should not, by themselves, be interpreted as psychotic behavior. For example, using a representative sample, Glenn and Cunningham found that talking out loud to one's self was very common in individuals with Down syndrome and approximated the typical developmental pattern. Self-talk is developmentally appropriate and is theorized to have important adaptive functions, particularly in relation to self-regulation when, for example, young children guide themselves through tasks by talking out loud to themselves (e.g., Mead, 1934).
A similar issue arises with compulsive, ritualistic-type behaviors, which are also reported as being common in children with Down syndrome (e.g., Evans et al., 1997) and associated with psychiatric problems in adulthood (e.g., McCarthy, 2004). Ritualistic behaviors, however, have much in common with the routinized and compulsive-like behaviors that many researchers have concluded are part of typical development in young children. For example, Gesell, Ames, and Ilg (1974) observed that at around 2 to 3 years of age, children become highly compulsive, and their actions often have to be carried out in a particular way (e.g., they may have elaborate bed-time routines, expect things to be done “just right,” and may have particular food preferences). Routinized and compulsive behaviors may also be characterized by “insistence on sameness” and repetition (e.g., request to hear the same story, must have a particular object(s), watch the same video over and over, or carry out the same actions during play). When the routine is thwarted, children often show strong emotional reactions.
For older children and adults, researchers have suggested that routinized and compulsive behaviors are similar to, and probably part of, obsessive-compulsive disorder—OCD (Evans et al., 1997; Leonard, Goldberg, Rapaport, Cheslow, & Swedo, 1990). This raises the question of whether the high levels of routinized behavior seen in adults with Down syndrome are a reflection of typical development or indicative of psychiatric problems, including OCD.
There are key differences between routinized and compulsive behaviors and OCD behaviors. Obsessions (recurrent, distressing unwanted thoughts) and compulsions (repetitive, ritualistic behaviors) are characteristic of OCD and restrict a child's development and have a negative impact on life-style. In contrast, many researchers have argued that routinized and compulsive behaviors have a positive and adaptive function for the developing child. For example, childhood routines enhance socialization and simplify complex situations (e.g., in children's repeated play patterns and games); they help to reduce and master anxiety, such as in bedtime routines involving familiar sequences of actions on the part of the parent and child (e.g., Leonard et al., 1990). Routines also help to develop a sense of security and control over the environment (Zohar & Felz, 2001). Thus, routines (and familiarity) increase the child's feeling of competence, reduce anxiety, and free-up cognitive capacity, all of which foster the likelihood that the child will explore and learn new ideas and skills.
Behaviors such as “just right” (e.g., when the child requires everything in their bedroom to be in place; or lines cars up in order of size; or insists on the same plate, cup, or food) are also functional for understanding order in the world and the developing capacity for classification (Piaget, 1953). The universality of such behaviors in early childhood strongly indicates that they have an adaptive function related to cognitive and social–emotional development (Evans et al., 1997). The question is whether routinized and compulsive behaviors have similar functions for older individuals with limited cognitive capacities. If not, when and how do they become a symptom of pathology?
In order to investigate such behaviors more systematically, Evans et al. (1997) devised a standardized measure of routinized and compulsive behaviors using a 19-item parent-report questionnaire (The Childhood Routines Inventory). They found that parents of 62% of 2- to 3-year olds (range = 38% to 83%) rated them as showing routinized and compulsive behaviors, decreasing to 49% (range = 17% to 84%) by around 6 years of age. They identified two factors: just right behaviors and repetitive behaviors, mirroring the types of behaviors discussed above. Evans et al. concluded that routinized and compulsive behaviors are part of typical child behavior and, therefore, do have an adaptive function, but they decline in frequency at around 6 years of age. This raises a similar question to the one above: If routinized and compulsive behaviors do not decline, do they become maladaptive?
Using the Childhood Routines Inventory, Evans and Gray (2000) compared 50 typically developing children matched for mental age (MA) with 50 children who had Down syndrome (mean chronological age [CA] 12 years, mean MA 4.9 years). They also asked the parents to complete the Child Behavior Checklist (Achenbach, 1991) and, by interview, the Vineland Adaptive Behavior Scales Screener—the Vineland (Sparrow, Carter, & Cicchetti, 1993). They hypothesized that total Childhood Routines Inventory scores would relate positively to adaptive behavior (Vineland screener ratings) in younger children, but not to those of older children, because routinized and compulsive behaviors are not developmentally appropriate at older ages. No support was found for this hypothesis for the younger children (under 5 years MA) with or without Down syndrome. However, for typically developing children with MAs over 5 years, the Vineland ratings did correlate significantly with just right scores, suggesting an adaptive function. In contrast, for both the younger and older MA children with Down syndrome, total Childhood Routines Inventory scores were correlated with problem behaviors (Child Behavior Checklist scores). The number (frequency) of routinized and compulsive behaviors was similar, suggesting typical developmental progress, but the behaviors were rated as significantly more intense for the children with Down syndrome. Evans and Gray concluded that routinized and compulsive behaviors may serve qualitatively different functions across the two groups of children and may indicate pathology in those with Down syndrome. Wigren and Hansen (2003) similarly found significantly more intense repetitive behavior on the Childhood Routines Inventory for 50 individuals with Prader-Willi syndrome (ages 5 to 18 years) compared to 50 typically developing 4-year-old children. Unfortunately, there was no matching for MA.
Evans and Gray (2000) also found that children with Down syndrome showed MA-related decreases in Childhood Routines Inventory scores similar to typically developing children. However, McCarthy (2004) found increasing ritualistic behaviors in adults with Down syndrome. The suggested similarity between routinized and compulsive behaviors and ritualistic behaviors suggests the need to investigate routinized and compulsive behaviors in older individuals with Down syndrome.
Because children and adults with intellectual disabilities have less capacity to learn, one would expect more repetitions and routines compared to typically developing children and that these would help learning. As noted earlier, routines require less cognitive capacity and give a sense of control, competence, and security. However, if such routines become compulsive to the point that the child is prevented from exploring new challenges and learning new things, then one would expect maladaptive effects, including lower scores on adaptive behavior, increasing behavior problems, and possibly increasing levels of OCD.
Thus, our aim in the present study was to investigate the functions of routinized and compulsive behaviors for children and young people with Down syndrome and any changes with age. In Study 1 we replicated the research of Evans and Gray (2000); and in Study 2, investigated the intensity, number, and functions of routinized and compulsive behaviors in young adults with Down syndrome.
Based on the results of Evans and Gray (2000), we hypothesised that (a) The intensity of routinized and compulsive behaviors would be significantly higher in children with Down syndrome than in typically developing children, although the number of routinized and compulsive behaviors would not be significantly different (Study 1). (b) The intensity and number of routinized and compulsive behaviors would decrease with MA (Studies 1 and 2). (c) Routinized and compulsive behaviors would be positively correlated with adaptive behavior in early childhood (or for lower MA) and positively correlated with behavior problems in later childhood (or for higher MA) (Studies 1 and 2). We also investigated any relationship between routinized and compulsive behaviors and OCD in the young adults with Down syndrome (Study 2).
Ethical approval was obtained from the University Ethics Committee. The young adults with Down syndrome were mostly recruited from the Manchester Down syndrome cohort (Cunningham, 1996, 2006). Others, and children with Down syndrome, were recruited from local Down Syndrome Associations plus a database of younger children who had participated in previous research (e.g., Glenn, Dayus, Cunningham, & Horgan, 2001). The typical comparison group was obtained from two preschools and two primary schools. One preschool and primary school were in an innercity area and included more children with less advantaged backgrounds; the others were from a small rural town, which had children from a range of socioeconomic backgrounds typical of Northern England.
This pool of participants consisted of 96 typically developing children (M CA 4.6 years, SD = 1.4, range = 1.5 to 8.3) and 118 children and young adults with Down syndrome (M CA 16.7 years. SD = 8.3, range = 2.0 to 29.3, M MA 5.1 years, SD = 2.4). All participants with Down syndrome had chromosomally verified trisomy 21.
We used the pool to select children and adults on the basis of MA and gender in order to form matched groups. Both the typically developing children and those with Down syndrome were divided into two groups. The younger group (MA 5 years or less) contained 26 children with Down syndrome (14 male, M MA 3.6 years, SD = 1.1, M CA 10.1 years, SD = 3.6) and 26 typically developing children (14 male, M MA 3.7 years, SD = 1.0, M CA 3.4 years, SD = 1.1). The older group (MAs over 5 years) contained 24 children and young people with Down syndrome (12 males, M MA 6.9 years, SD = 1.7, M CA 16.7 years, SD = 6.2) and 24 typically developing children (12 male, M MA 6.7 years, SD = 1.3, M CA 5.3 years, SD = 1.3).
The participants were 43 young adults (27 male, 16 female) with Down syndrome. Their mean CA was 25.4 years (SD = 1.8), and mean MA was 6.2 years (SD = 2.5).
The British Picture Vocabulary Scales (Dunn, Dunn, Whetton, & Burley, 1997) provided an index of developmental status. Although a measure of receptive vocabulary, it is often used to provide an MA for samples with intellectual disability because it does not require speech, covers an MA range from 2 to 16 years, and vocabulary subtests have proved to be the most important contributors to comprehensive tests of intelligence.
Repetitive and compulsive-like behavior
The Childhood Routines Inventory (Evans et al., 1997) consists of 19 behaviors that the parent rates from 1 (not present) to 5 (very common). Evans et al. reported a Cronbach's alpha of .89 and identified two factors: Just Right Behaviors (e.g., “prefers to have things done in a particular order or certain way”) and Repetitive Behaviors (e.g., “repeats certain actions over and over”), each including five items. Two scores are produced: the total number of compulsive and repetitive behaviors and the mean intensity (rated from 1 to 5) of such behaviors.
The Child Behavior Checklist (Achenbach, 1991) is a 112-item checklist on which parents rate child behavior problems from 0 to 2. Two versions were used: one for participants with an MA of 4 years and younger (Achenbach &Rescorla, 2000), and one for those with an MA over 4 years (Achenbach, 1991). Standard T scores were used in the analyses.
Hudziak et al. (2004) demonstrated that 8 items from the Child Behavior Checklist could be used as an index of obsessive-compulsive disorder; they named this the Child Behavior Checklist Obsessive-Compulsive Scale. Using raw scores, they reported that for a summed score of 5, the scale showed high sensitivity (92%), moderate specificity (67%), high negative predictive value (90%), and moderate positive predictive value (73%) in individuals who had been diagnosed as having Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) OCD by psychiatrists.
Vineland Adaptive Behavior Scales Screener—Vineland Screener (Sparrow et al., 1993). This brief version of the Vineland uses a semi-structured interview with parents. We assessed the child/young person's functioning in the domains of Communication, Daily Living, and Socialization.
Parents of children and young people received a letter in which we explained the research plus the Childhood Routines Inventory. They were asked to (a) complete the Childhood Routines Inventory and record the child or young person's age and gender, (b) state whether they would participate in the second part of the study requiring them to complete the Child Behavior Checklist, the Vineland Screener (by telephone interview), and give permission for their child to be assessed on the British Picture Vocabulary Scales. When possible, we also asked the children and young people for their consent.
Analysis of Data
We used MANOVA and ANOVA, as appropriate, to compare groups, with Box's M test, Bartlett's Test of Sphericity, and Wilk's exact F calculated to ensure that statistical assumptions were satisfied. Where significant results were obtained, we conducted further univariate analyses. We used Pearson product-moment correlations to explore associations between measures. Two-tailed tests were used throughout.
The two main issues were (a) whether there was a difference in number or intensity of routinized and compulsive behaviors for children with and without Down syndrome and (b) whether routinized and compulsive behaviors decreased with age.
Table 1 presents the mean (SD) number and intensity scores on Total Childhood Routines Inventory, just right behaviors, and repetitive behaviors for children with Down syndrome and typically developing children.
Number of routinized and compulsive behaviors
There were no significant differences between children with and without Down syndrome for the number of routinized and compulsive behaviors rated by parents, F(3, 94) = 2.05, p = .113, effect size .061, power = .51, nor were there any significant differences between those with MAs below and above 5 years, F(3, 94) = .70, p = .554, effect size = .02, power = .19; or interaction between child group and MA level, F(3, 94) = 2.53, p = .062, effect size = .075, power = .61.
Intensity of routinized and compulsive behaviors
The group with Down syndrome showed significantly higher intensity of Total Childhood Routines Inventory behaviors than did the typically developing children, F(1, 96) = 4.19, p = .044, effect size = .044, power = .53 and for repetitive behaviors, F(1, 96) = 8.92, p = .004, effect size = .09, power = .84, but not just right behaviors, F(1, 96) = 3.01, p = .086, effect size = .03, power = .40. There was no significant effect of MA level on intensity of any routinized behaviors, F(3, 94) = 1.59, p = .197, effect size = .05, power = .41, or interaction between child group and MA level, F(3, 94) = .90, p = .447, effect size = .03, power = .24.
Thus, and as expected, we found that the group with Down syndrome had significantly more intense routinized and compulsive behaviors than did typically developing children, but numbers of such behaviors were not different. We did not find the expected decrease in routinized and compulsive behaviors (either number or intensity) with increasing MA in those with or without Down syndrome.
As can be seen in Table 2, the parents of children with Down syndrome rated them as having significantly more behavior problems than did the parents of the typically developing children, ANOVA: F(1, 96) = 35.7, p = .000, effect size .271, power 1.000. There was no significant effect of MA on behavior problems, F(1, 96) = .004, p = .952, effect size .000, power .05, nor any interaction between child group and MA, F(1, 96) = 1.37, p = .271, effect size .271, power = .212.
The third issue concerned the relationships between routinized and compulsive behaviors and adaptive behavior and behavior problems. Correlations were computed between all the variables. Only those that were significant are reported.
Adaptive behavior was indexed by the Vineland and MA scores. For typically developing children with MAs of 5 years or less, there were significant correlations between MA and the intensity of total, r = .57, p = .003, just right, r = .56, p = .003, and repetitive, r = .48, p = .013, Childhood Routines Inventory items. There were also significant correlations between intensity of Total Childhood Routines Inventory and Vineland scores, r = .39, p = .046, and between number of Childhood Routines Inventory items rated as present and MA, r = .50, p = .010. For the children with Down syndrome with MAs of 5 years or less, there were significant correlations between the number (but not intensity) of Total Childhood Routines Inventory items, and just right items and the Vineland score, r = .40, p = .044; r = .44, p = .024, respectively.
For both typically developing children and those with Down syndrome with MAs more than 5 years, there were no significant correlations between routinized and compulsive behaviors and adaptive behavior; in fact, the correlations between Childhood Routines Inventory and adaptive behavior were negative.
Thus, for both typically developing children and those with Down syndrome under 5 years of developmental age, there was evidence that routinized and compulsive behaviors were adaptive, in so far as children showing more or more intense routinized and compulsive behaviors had significantly higher MAs or adaptive behavior. However, there was no significant relationship for those over 5 years of developmental age.
Maladaptive behavior (Child Behavior Checklist scores)
For MAs of 5 years and less, for both typically developing children and those with Down syndrome, there were no significant correlations between behavior problems and either number or intensity of Childhood Routines Inventory items. This again supports the idea that at this age routinized and compulsive behaviors have no negative impact.
For MAs over 5 years for the group with Down syndrome, it was the intensity (not the number) of total, r = .51, p = .012, and repetitive, r = .46, p = .026, Childhood Routines Inventory items that correlated with behavior problems. For the typically developing children with MAs over 5 years, the only significant correlation was between the number of repetitive behaviors and behavior problems, r = .42, p = .040.
Thus, for those with MAs over 5 years, there was evidence of a link with behavior problems. For participants with Down syndrome, the more intense the routinized and compulsive behaviors the more behavior problems were rated by parents. For the typically developing children, the link was less with only some indication of more repetitive behaviors being associated with more problem behaviors.
The fourth issue was based on the assumption that the pattern of typical development and routinized and compulsive behaviors can be applied to children with Down syndrome. If correct, we would expect that the intensity and number of routinized and compulsive behaviors would decrease with developmental age. Of the 46 young adults in the study, only 12 had an MA less than or equal to 5 years, which reduced the likelihood of significant correlations.
Table 3 presents the intensity and number of Childhood Routines Inventory items, behavior problem ratings, and OCD scores for the 43 young adults with Down syndrome. A MANOVA showed no significant effect of MA on Childhood Routines Inventory scores: intensity: F(3, 39) = 2.01, p = .129; number: F(3, 39) = 2.44, p = .079, and, therefore, there was no indication of a decline with increasing MA.
Using the same assumptions as those in Study 1, we expected that routinized and compulsive behaviors would be positively correlated with adaptive behavior in young adults with Down syndrome with MAs equal or less than 5 years and positively correlated with behavior problems in those with MAs over 5 years
Adaptive behavior (Vineland and MA scores)
For both MA groups (≤ 5 years and > 5 years), there were no significant correlations between any Childhood Routines Inventory scores and adaptive behavior. In fact, these correlations were mostly negative.
Maladaptive behavior (Child Behavior Checklist scores)
For those with MA less than 5 years, 5 out of 6 Childhood Routines Inventory scores correlated significantly with the T score from the Child Behavior Checklist (intensity: Total Childhood Routines Inventory, r = .72, p = .012; just right, r = .62, p = .044; repetitive, r = .63, p = .038. For number: Total Childhood Routines Inventory, r = .67, p = .023; just right: r = .56, p = .074; repetitive, r = .71, p = .014. For those with MA greater than 5 years, the intensity (but not the number) of all Childhood Routines Inventory scores correlated significantly with the T score from the Child Behavior Checklist: Total, r = .51, p = .004; just right, r = .37, p = .047; repetitive, r = .37, p = .045.
Thus, as with Study 1, there was no indication of decline in either number or intensity of routinized and compulsive behaviors with increasing developmental age. Furthermore, we found no evidence that routinized and compulsive behaviors were adaptive for young adults with Down syndrome, irrespective of MA. With respect to behavior problems, significant associations were found for those with MAs more than 5 years, but not for those with MAs less than 5 years. However, for the latter the correlations were of the same order, but not significant, possibly due to small number in the group.
Further exploration of the data
Because we had not found any decrease in routinized and compulsive behaviors with increasing MA in Studies 1 and 2, we decided to explore any changes in such behaviors in relation to CA or MA with the total pool of participants. Childhood Routines Inventory scores were available for 96 typically developing children and 118 individuals with Down syndrome.
The typically developing children were divided into 4 CA groups based on quartiles (Ms: 36, 50, 61, and 76 months). In the same way, individuals with Down syndrome were divided into 4 quartiles using MA (Ms 27, 48, 69, and 100 months). This produced similar numbers of participants in each quartile. The mean scores and SDs for each group were very similar and showed no decline in number or intensity of routinized and compulsive behaviors with CA for the typically developing children, or with MA for the individuals with Down syndrome. For example, the intensity of Childhood Routines Inventory varied from a minimum mean score of 2.3 (.87) at mean 3 years of age to 2.6 (.91) at mean 6.5 years for typically developing children, and from a minimum of 2.6 (1.1) at mean MA 5.8 years to 2.7 at the other three MAs for the children and young people with Down syndrome. Thus, again, we failed to find evidence to support the suggestion in previous research of a decline around 5 years of age.
The fifth issue
Was there any relationship between OCD and routinized and compulsive behaviors for young adults with Down syndrome? For the OCD index (Child Behavior Checklist Obsessive-Compulsive Scale) in the younger MA group, there were positive correlations between Childhood Routines Inventory scores and the OCD Index, but none of these were significant. For those with MAs more than 5 years, there were significant correlations with the OCD Index (Intensity: Total Childhood Routines Inventory, r = .56, p = .001; just right, r = .41, p = .024; repetitive, r = .48, p = .007; and number Total Childhood Routines Inventory, r = .36, p = .050, suggesting that there may be a relationship between OCD and routinized and compulsive behaviors.
The discussion of the results and all conclusions are subject to the caveat that the study was based on parental reports rather than direct and independent observation. This raises the possibility of bias; what is seen as acceptable behavior in the younger child may be viewed as unacceptable or problematic in older children. Conversely, parents may expect different behavior as characteristic of the syndrome and so be more aware, or more tolerant. Thus, some caution is needed in forming conclusions.
Number and Intensity of Routinized and Compulsive Behaviors
The parents of both MA groups with Down syndrome reported significantly more intense routinized and compulsive behaviors than did the parents of typically developing children, but the same number of routinized and compulsive behaviors. The latter finding suggests that the range and nature of the behaviors is similar to that of typical children. Thus, like many other developmental sequences, children with Down syndrome show the same developmental organization for routinized and compulsive behaviors as do typical children, albeit delayed (Zigler & Hodapp, 1986). Given the limited repertoire and slow progress of children with Down syndrome, and heightened parental engagement and concern with their learning, it may not be surprising that parents rate routinized and compulsive behaviors as more intense. However, this may not just be a bias. The higher levels of intensity are probably closely related to the limited capacity for learning, due to the commonly described characteristics of intellectual disability, such as slower information-processing, need for more repetition and practice to consolidate learning, attentional deficits, and increased rigidity (e.g., Kounin, 1941; Zigler & Hodapp, 1986). Thus, frequent repetition and insistence on sameness may have a positive function by compensating for intellectual disability. As discussed later, routinized and compulsive behaviors in younger children do appear to be positively associated with developmental progress. Thus far, these results support those of Evans and Gray (2000).
Routinized and Compulsive Behaviors and Change With Age
In the present study we did not find a decrease between the under and over 5 years groups nor any strong suggestion of a decrease associated with CA or MA for either the typically developing children or the children and adults with Down syndrome. This was contrary to our initial expectations and the results of Evans and Gray (2000). This adds to the current debate about whether routinized and compulsive behaviors continue after 5 to 6 years of age. Evans et al. (1997) did not assess children older than 6, and, therefore, the decline they described may not be characteristic. Others have suggested that routinized and compulsive behaviors may continue in older children but be different in kind to those displayed earlier in childhood. Leonard et al. (1990), for example, noted that older children often start collections of objects, become superstitious about stepping on cracks, or may become “obsessed” with a rock star or a computer game. Zohar and Bruno (1997) reported that obsessive-compulsive type behaviors (assessed by the Maudsley Obsessive Compulsive Inventory—Hodgson & Rachman, 1977) were still present in children aged 8 years and then decreased significantly up to 14 years of age. Taken with the present results, these researchers' findings suggest that routinized and compulsive behaviors are still present in later childhood after 6 years of age. The mean MA reported from several studies of young adults with Down syndrome was between 5 and 6 years and the majority fell between 4 and 8 years (e.g., Cunningham, 2006). Thus, the presence of routinized and compulsive behaviors would be expected from a delayed development perspective. If this were the case, and taking into account limited behavioral repertoires and slower learning, it would be advisable to examine the nature and functions of routinized and compulsive behaviors more closely before assuming they are maladaptive.
Given the worries about the persistence of routinized and compulsive behaviors in adults with Down syndrome, a study of older age groups of typically developing children, using the Childhood Routines Inventory, is needed. As well as looking for decline in routinized and compulsive behaviors, researchers need to examine the nature of routinized and compulsive behaviors, probably with more exact descriptions of how parents interpret the items, what behavior parents are rating, and when they increase in intensity.
Functions of Routinized and Compulsive Behaviors
The key issue concerned the possible functions of routinized and compulsive behaviors and whether these behaviors change with development. For the younger participants with an MA of 5 years or less, both with and without Down syndrome, routinized and compulsive behaviors were found to be adaptive (i.e., they correlated positively and significantly with the Vineland and MA scores). This was particularly the case for just right behaviors, which are probably associated with the child's developing capacity for classification, order, feeling control over the environment, and press for self-determination. However, for the young adults with Down syndrome with MAs less than or equal to 5 years, there were no positive associations between Childhood Routines Inventory scores and MA or Vineland scores, suggesting that the adaptive function seen in younger children is no longer present, irrespective of MA levels. Caution is needed in this interpretation because we only had 12 young adults in this group. Their mean MA was 3.17 years, so all fell in the severe or profound categories for intellectual disability. It may be that their reduced capacity for learning acts as a barrier to any beneficial effects of routinized and compulsive behaviors. The function of routinized and compulsive behaviors for these young adults may have become compensatory, for example, to stimulate the nervous system, prevent it from being overwhelmed by complex external inputs, or support their functioning in a familiar environmental context.
The notion of compensatory mechanisms is partly supported by the finding of differing associations between routinized and compulsive behaviors and problem behaviors at different MAs and CAs for the participants with Down syndrome. Like Evans and Gray (2000), we found significant correlations between routinized and compulsive behaviors (particularly repetitive behaviors) and behavior problems for younger individuals with Down syndrome with MAs over 5 years. Similarly, for all the young adults in Study 2, there were significant positive correlations between routinized and compulsive behaviors and behavior problems. What we cannot determine from this data is any causal association between these two types of behavior. The high intensity of routinized and compulsive behaviors in older MA children and young adults with Down syndrome no longer has a positive function in relation to increasing MA (where a ceiling may have been reached), but may have different functions. As noted above, routinized and compulsive behaviors may be functional (e.g., in providing feelings of security for young people experiencing difficulties in coping with new situations in everyday life).
If we accept that routinized and compulsive behaviors are adaptive or compensatory, we should question the notion of any equation with OCD. Only 2 people (5%) were over the threshold of 5 points suggested by Hudziack et al. (2004) on the OCD measure, and neither of these individuals had a psychiatric diagnosis of OCD, which has a relatively low incidence in individuals with Down syndrome compared to other conditions associated with intellectual disability and is reported to be between 0.8 to 4.5% in this population (Prasher, 1995). These results suggest a distinction between routinized and compulsive behaviors and OCD and question a simplistic view of any continuity between the two sets of behavior. We also question the validity of the Child Behavior Checklist Obsessive-Compulsive Scale for young people with Down syndrome, because obsessions are concerned with internal thoughts and worries rather than overt behavior, and it is compulsive behavior rather than obsessive thoughts that is most common in this population.
Overall, therefore, the findings suggest that routinized and compulsive-like behavior is typical in the early developmental years for children with and those without Down syndrome. Parents of young children with Down syndrome may be reassured that routinized and compulsive behaviors, in particular just right behaviors, are developmentally appropriate and that repetition is necessary for learning to become consolidated. This assertion implies that parents and teachers should be cautious before deciding that the behaviors are detrimental to development and everyday life. Rather than inhibiting or attempting to extinguish the behavior, it may be preferably to examine its function for the child or young adult and then build upon this to extend their repertoire of skills and understanding.
Thus, the question of whether routinized and compulsive-like behaviors are typical or pathological appears too simple. We need to examine the nature, levels, and functions of these behaviors at different developmental and CAs. The data show that with increasing years (developmental and chronological), parental ratings of routinized and compulsive behaviors become associated with ratings of behavior problems for both typical children and for children and young people with Down syndrome. What is not clear is whether this is adaptive or compensatory or indicative of possible pathology. Use of the term compulsive-like may overemphasise pathology, and we found little support for a direct relationship between routinized and compulsive behaviors and OCD or other psychiatric conditions. Routinized and compulsive behaviors mostly involve routines of behavior rather than true compulsions, which highlights a distinction that should be made between repetition (necessary for learning) and actions that appear repetitive and not leading to new skills. The issue about some of these repetitions (e.g., repeated play with the same well-mastered jigsaw, insistence on hearing/reading the same story or viewing the same video) is that more competent observers may not appreciate how much information is still being learned or experienced as novel by the child. There are also the feelings of security and sense of mastery associated with the familiar. Faced with uncertainty and associated anxiety, we all tend to revert to familiar knowledge or actions. We establish familiar routines to cope with a complex world. People with intellectual disabilities will rely more on these routines and require more repetitions for learning to take place.
If, after careful examination, the routine-compulsive type behavior seems to be interfering with the person's everyday life, then a first step would be to examine its function for the person and possible reasons for its maintenance. Behavioral techniques could then be used to modify the behavior. This should be done before making any diagnosis of personality disorders or OCD.
In conclusion, we argue that the issue of the possible pathological nature of more intense routine and compulsive-like behaviors in individuals with Down syndrome is worthy of further investigation. In particular, we need research (clinical and empirical) that describes the repetitive and ritualistic-like behaviors more precisely and in relation to developmental level, context, and possible function for children and adults. The Childhood Routines Inventory also needs further standardization using older typically developing children, individuals with intellectual disability from other causes, and those with behavior problems or diagnosed psychiatric conditions.
A grant from the Nuffield Foundation supported the study. We thank the children, young people, and families who participated. We also thank Linda Mason and Helen Poole for assistance in data-collection.
Sheila Glenn, PhD, Professor (firstname.lastname@example.org) and Cliff Cunningham, PhD, Professor, Faculty of Health and Applied Social Sciences, Liverpool John Moores University, Kingsway House, Hatton Garden, Liverpool L3 2AJ, United Kingdom