Relationship of hypersensitivity to anxiety and depression in children with high-functioning pervasive developmental disorders
Hisashi Tsuji, MD, Department of Neuropsychiatry, Osaka City University Graduate School of Medicine 1-4-3 Asahimachi, Abenoku, Osaka 545-8585, Japan. Email: email@example.com
Aims: Sensory–perceptual abnormalities, which include hyper- and hyposensitivity, have been identified by numerous researchers as prevalent in individuals with pervasive developmental disorders (PDD). Hypersensitivity has a greater impact on PDD patients' daily lives than hyposensitivity. The purpose of the present study was to clarify the relationship of hypersensitivity to anxiety, depression and other psychopathology in children with PDD.
Methods: Sixty-four children were divided into a hypersensitivity group (HG; n = 43) and a non-hypersensitivity group (non-HG; n = 21), and compared for anxiety, depression and other psychopathology on the Child Behavior Checklist (CBCL), State–Trait Anxiety Inventory for Children (STAIC) and Children's Depression Inventory (CDI).
Results: The HG group had significantly higher scores than the non-HG group in Total, Internalizing, and Somatic complaints on the CBCL. On STAIC, the mean sore of Total Score, State Score and Trait Score in the HG group tended to be higher than in the non-HG group, but the difference was not significant. The score on the CDI in the HG group was significantly higher than that in the non-HG group.
Conclusion: PDD children with hypersensitivity have more serious psychopathologies, especially internalizing symptoms including depression.
PERVASIVE DEVELOPMENTAL DISORDERS (PDD) are characterized by deficits in social interaction, impaired communication, and repetitive and stereotyped patterns of behavior, interest and activities. They are most frequently diagnosed on the basis of behavioral criteria described in DSM-IV.1 But these are not the only characteristics of the disorder: an unusual pattern of sensory perception and reaction have also long been associated with PDD.2 Sensory–perceptual abnormalities have been reported by numerous researchers as prevalent in individuals with PDD.3–16 Physiologically, experimental studies of sensory–perceptual phenomena suggest an usual pattern of responsiveness to sensory stimuli in individuals with PDD.17,18 Reports of sensory abnormalities in PDD are found across all sensory domains.10,17–20 These include hypersensitivity and hyposensitivity to sensory stimuli. For example, in the tactile domain, parents frequently report their autistic child's problems in adapting to new types of clothing fabric. In the auditory domain, some individuals with autism report being frightened by every ring of a telephone. In the visual domain, some children become sick at the sight of branches waving in the breeze.
Royeen and Lane suggested the term ‘sensory defensiveness’ as a modulation disorder in the processing of sensory input by the central nervous system. Sensory defensiveness is defined as a tending to react negatively or with alarm to sensory input that is generally considered harmless or non-irritating.21 It is also characterized by hypersensitivity or overresponsiveness.
Kinnealey and Fuiek found anxiety and depression levels to be elevated in a population of adults with sensory defensiveness but who were free of other psychopathology.22 Pfeiffer et al. found a relationship for hypersensitivity with both anxiety and depression and also for hyposensitivity with depression using the parent-report instruments in children and adolescents with Asperger's disorder.9 In PDD children, however, subjective sensory–perceptual abnormalities and psychophysical impairments have not been well characterized.
Investigation of the entire scope of sensory–perceptual abnormalities may be beneficial for understanding the range of unusual patterns of sensory perception and reactions, but there may also be a need to separately examine the features of hypersensitivity and hyposensitivity.
By considering hypersensitivity and hyposensitivity as separate aspects of sensory–perceptual abnormalities, researchers may be better able to understand sensory-related behavioral patterns as a whole.23 We supposed that hypersensitivity had a greater impact on PDD patients' daily lives than hyposensitivity. The present study focused on the hypersensitivity form of sensory–perceptual abnormalities, and was carried out to clarify the relationship of hypersensitivity to anxiety, depression and other psychopathology in children with high-functioning PDD (HFPDD).
The subjects were 64 HFPDD children. They were selected from 471 consecutively referred children aged 6–15 years (elementary or junior high school students) receiving treatment at the children psychiatry outpatient clinic of the Osaka City University Hospital between January 2003 and January 2007.
At the outpatient clinic, diagnosis of the children was established after extensive diagnostic procedures. We followed strict diagnostic guidelines for PDD by (i) interviewing parents about their children's developmental history and any history of autistic behavior that their children exhibited since infancy via the questionnaires; (ii) obtaining information from teachers via questionnaires about PDD; (iii) having the interviews carried out by more than two child psychiatrists who had experience with PDD examination and treatment, and (iv) observing the child's clinical course for more than 3 months. In the present study, to investigate subjective psychophysical impairments, children who had mental retardation (IQ < 70 on the Wechsler Intelligence Scale for Children–Third Edition [WISC-III]) were excluded. Children with acute psychotic state, cerebral palsy, or epilepsy were also excluded.
Regarding participation in the study, we explained the investigational purpose, the procedures, the potential risks and their alternatives to participation, and obtained written consent from the subject and the parents of each subject.
The inclusion criteria for PDD included autistic disorder, Asperger's disorder and PDD not otherwise specified (PDD-NOS) diagnosed using DSM-IV criteria. In addition, because PDD-NOS is defined as a residual category of PDD and has no operational criteria,1 we used the Buitelaar and van der Gaag diagnostic criteria for PDD-NOS, which require a total of three or more items from criterion A of DSM-IV criteria for autistic disorder, including at least one item from (1) qualitative impairments in social interaction.24 As a result, 64 children were diagnosed as having PDD, consisting of 51 boys and 13 girls. Of these, 57 children were diagnosed as having autistic disorder and seven as having PDD-NOS. There were no children diagnosed with Asperger's disorder. The mean age of the subjects was 10.0 years. The mean full-scale IQ score on the WISC-III was 96.6 (Table 1).
Table 1. PDD subject characteristics (mean ± SD)
|Age (years)||10.0 ± 2.4||9.6 ± 2.4||10.6 ± 2.5||−1.472||0.146|
|Male/Female (n)||51/13||31/12||20/1|| ||0.045†|
|Full IQ||96.6 ± 12.1||95.9 ± 12.2||97.9 ± 12.3||−0.573||0.569|
|Verbal IQ||98.0 ± 14.8||97.0 ± 14.4||100.2 ± 15.7||−0.752||0.455|
|Performance IQ||96.2 ± 12.1||96.7 ± 12.4||95.1 ± 11.7||0.458||0.649|
There are no widely agreed criteria and rating scale for hypersensitivity that have sufficient reliability and validity to measure hypersensitivity. Therefore, we defined hypersensitivity as a condition of extreme discomfort or irritability in response to non-noxious stimulation in a typical individual. The children and their parents were independently interviewed to obtain information about the child's hypersensitivity in the auditory, visual, tactile, olfactory, and gustatory domains. Information about the child's hypersensitivity was obtained from their teacher. Based on this information, two child psychiatrists classified the hypersensitivity group (HG) as subjects with hypersensitivity in one or more sensory modalities, and the non-hypersensitivity group (non-HG) as subjects without any hypersensitivity. Of the 64 PDD children, 43 (67%) were classified into the HG group, and 21 (33%) into the non-HG group. Anxiety, depression and other psychopathology of the HG group were compared with those of the non-HG group.
Child Behavior Checklist
Itani et al. standardized a Japanese version of the Child Behavior Checklist (CBCL) developed by Achenbach et al.25 This consists of a 113-item questionnaire rated by their parents (CBCL) using the following scale: 0, not true (as far as you know); 1, somewhat or sometimes true; and 2, very true or often true. The items describe behavioral and emotional problems: Withdrawn, Somatic Complaints, Anxious/Depressed, Social Problems, Thought Problems, Attention Problems, Delinquent Behavior, and Aggressive Behavior. Raw scores are converted into T-scores (mean 50; SD 10) which are adjusted for age and gender.
State–Trait Anxiety Inventory for Children
The State–Trait Anxiety Inventory for Children (STAIC) is a self-reporting questionnaire that is widely used to measure anxiety in children.26 It consists of two separate 20-question rating scales, one for ‘state’ anxiety (acute, transitory) and the other for ‘trait’ anxiety (chronic, pervasive). The Japanese version was standardized and has satisfactory internal reliability.27 The STAIC was administered to subjects ≥10 years of age.
Children's Depression Inventory
The Children's Depression Inventory (CDI) is one of the most widely used self-reporting questionnaires to measure depressive symptoms in children and adolescents.28 It has 27 items, each of which consists of three statements on a 3-point scale (ranging from 0, not true; to 3, very true). The total CDI score ranges from 0 to 54. The Japanese version was translated by Murata et al. and has satisfactory internal consistency reliability.29,30 Murata et al. recommended a cut-off for clinical depression of 22.29 The CDI was used for subjects ≥8 years of age.
All statistical analysis were performed using SPSS 11.5J for Windows (SPSS Japan, Tokyo, Japan). We used Fisher's exact test for categorical comparisons of the data. For continuous variables, the normality of the distribution of scores on all measures was examined. When the data were normally distributed, t-tests were used to examine group differences. Mann–Whitney U-test was used when the data were not normally distributed. Statistical significance was set at P < 0.05 (two-tailed).
Table 1 shows the age, sex distribution and IQ of the subjects in the HG and non-HG groups. There were no significant differences in age, full IQ, verbal IQ and performance IQ between the two groups. The rate of female subjects having hypersensitivity was significantly higher than that of male subjects. The mean age of girls was significantly lower than that of boys (8.6 ± 2.3 years vs 10.3 ± 2.4 years, respectively: t = −2.353, P = 0.022).
Table 2 lists the presence of hypersensitivity in each modality. The most prevalent hypersensitivity was in the auditory domain (n = 35). Twenty-one children had hypersensitivity in one modality, 18 children had this in two modalities, and four children had hypersensitivity in three or more modalities.
Table 2. Hypersensitivity in subjects with PDD
Table 3 lists T-scores for the CBCL. The HG group had significantly higher T-scores than the non-HG group on Total, Internalizing and Somatic complaints on the CBCL.
Table 3. T-score of CBCL vs PDD subgroup
|CBCL||n = 63||n = 43||n = 20|| || |
|Total||70.9 ± 8.5||72.5 ± 9.4||67.5 ± 5.1||2.750||0.008|
|Internalizing||67.6 ± 9.6||69.4 ± 10.3||63.6 ± 6.9||2.313||0.024|
|Externalizing||67.7 ± 10.8||68.6 ± 11.3||65.5 ± 9.8||1.049||0.298|
|Withdrawn||66.3 ± 12.3||66.9 ± 14.0||64.6 ± 7.9||0.688||0.494|
|Somatic Complaints||60.1 ± 9.7||61.6 ± 10.4||56.7 ± 7.0||2.198||0.032|
|Anxious/Depressed||65.6 ± 9.8||67.2 ± 10.2||62.3 ± 8.5||1.862||0.067|
|Social Problems||70.5 ± 9.1||70.4 ± 9.4||70.3 ± 8.5||0.049||0.961|
|Thought Problems||63.8 ± 11.5||65.0 ± 11.4||61.5 ± 11.7||1.148||0.255|
|Attention Problems||70.4 ± 7.3||71.4 ± 7.9||68.3 ± 5.6||1.829||0.073|
|Delinquent Behavior||63.8 ± 9.2||64.1 ± 9.2||63.6 ± 9.5||−0.215||0.830|
|Aggressive Behavior||67.1 ± 10.6||68.0 ± 11.1||64.5 ± 9.2||1.241||0.220|
Table 4 lists the scores on the STAIC and the CDI. On the STAIC there was no significant difference in the Total score, State score and Trait score between the HG and the non-HG groups. The score on the CDI in the HG group was significantly higher than that in the non-HG group.
Table 4. STAIC and CDI score vs PDD subgroup
|STAIC||n = 18||n = 12||n = 6|| || |
|Total score||70.4 ± 20.1||75.8 ± 20.2||59.8 ± 17.1||17.5||0.083|
|State score||33.6 ± 12.6||36.7 ± 13.3||27.3 ± 8.8||22.5||0.201|
|Trait score||36.9 ± 9.6||39.1 ± 8.5||32.5 ± 11.0||22.0||0.189|
|CDI||n = 21||n = 15||n = 6|| || |
|19.0 ± 8.6||21.5 ± 8.6||12.8 ± 5.1||2.283||0.034|
This is the first study to investigate the relationship of hypersensitivity to anxiety and depression and other psychopathology in Japan.
A number of studies provide evidence to support the view that unusual sensory responses are present in the majority of children with PDD. Some investigators have suggested that this rate is closer to 100% in children with PDD.10–12 These estimates do not, however, discriminate between over- and underresponsiveness. In the present study the rate of subjects with hypersensitivity was 67%. Despite the different cultural background and the different level of intelligence, as with the present study, the aforementioned studies found high prevalence of hypersensitivity in PDD. These findings suggest that sensory hypersensitivity is highly prevalent in PDD. In the present study, hypersensitivity in the auditory domain was most common. This is consistent with previous findings.3,10,13,14,16
Sex differences in hypersensitivity have been poorly studied. In the present study the number of girls with hypersensitivity was significantly higher than that of boys. The mean age, however, of the girls was significantly lower than that of the boys. A confounding effect of aspects of maturation may account for the differences. Holtman et al. found that female subjects with HFPDD had more severe social and attention problems than males.31 In their study, however, there is no information regarding hypersensitivity. The severity of social and attention problems in HFPDD female subjects might be affected by sensory hypersensitivity.
Internalizing score, Total score and Somatic Complaints score on the CBCL were significantly higher in the HG than in the non-HG subjects. This suggests that HG subjects had more serious psychopathologies, especially internalizing symptoms.
The mean CDI score was significantly higher in the HG than in the non-HG subjects. On STAIC, the mean Total score, State score and Trait score in the HG group tended to be higher than in the non-HG group, but the difference was not significant. Kinnealey and Fuiek have identified the relationship between sensory defensiveness and levels of anxiety and depression in adults diagnosed with no medical or mental health issues.22 Pfeiffer et al. found a relationship for hypersensitivity with both anxiety and depression using the parent-report instruments in children and adolescents with Asperger's disorder.9 The present result is consistent with the findings of Pfeiffer et al. and Kinnealey and Fuiek.9,22 There is some evidence to suggest that an unusual pattern of sensory perception and reaction manifest themselves very early in development.14,15 The present results therefore suggest that hypersensitivity might be a possible risk factor for anxiety and depression. In contrast, some researchers have suggested that children and adolescents with PDD often have difficulty expressing and understanding their feelings and emotions.13,32 Thus the possibility of cognitive bias of PDD children should be taken into account, especially on the CDI, owing to the younger age range of subjects, given that some of them had a mental age under the age range. Also, the present sample size was relatively small. More research with a larger sample and using parent-report instruments to determine children's anxiety and depression together is needed. No significant difference was found in Anxious/Depressed score on CBCL. We believe, however, that CBCL is a less specific scale to measure anxiety and depression than the STAIC and CDI.
Tshe present results must be interpreted in the context of methodological limitations. These findings, based on a clinical population, should be interpreted with caution, because they may not generalize to other populations. The present results cannot be generalized to PDD with mental retardation, because the study involved only HFPDD subjects (IQ > 70). In the present study the reliability and validity of the criteria for hypersensitivity are not yet sufficiently established. Moreover, we did not measure the severity of hypersensitivity. In future studies operational criteria or a rating scale to measure the severity of hypersensitivity should be established. In addition, we did not analyze these data according to gender because of the relatively small sample size. More research with a larger sample is needed to investigate sex differences in hypersensitivity.
CLINICAL IMPLICATION AND CONCLUSIONS
The present findings suggest that HFPDD children with hypersensitivity have more serious psychopathologies, especially internalizing symptoms including depression. Thus, it is important to assess hypersensitivity in children with PDD. There may be potential to decrease internalizing symptoms including depression and improve quality of life using a treatment strategy with due consideration for hypersensitivity.