SEARCH

SEARCH BY CITATION

Keywords:

  • autism;
  • behavioral problems;
  • parenting style;
  • siblings

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

Aims:  The purpose of the present study was to investigate the behavioral problems and parenting style among children with autism and their siblings in an ethnic Chinese population.

Methods:  A total of 151 children with DSM-IV autistic disorder, aged 3–12, 134 siblings without autism, and 113 normally developing controls were recruited. Both parents reported their parenting styles and psychological status and mothers also reported children's behavioral problems.

Results:  Children with autism had significantly more severe behavioral problems and obtained less affection and more overprotection and authoritarian controlling from their parents than the other two groups. Compared to the controls, unaffected siblings showed some behavioral problems, and obtained less maternal care. Withdrawal and attention, social, and thought problems were the most associated behavioral syndromes to distinguish children with autism from those without.

Conclusions:  In addition to children with autism, who have a wide range of behavioral problems and impaired parent–child interactions, their siblings may be at risk for such problems.

AUTISM IS A pervasive, multi-factorial, highly heritable, clinically heterogeneous neurodevelopmental disorder with prominent impairments in social reciprocity, language impairment and restricted repetitive behaviors or interests.1 Two recent population cohort surveys in the UK reported that the prevalence of autism spectrum disorders (ASD) at age 9–10 years was 116.1 per 10 000 (childhood autism, 38.9 per 10 000; other ASD, 77.2 per 10 000),2 which is higher than the 58.7 per 10 000 in children aged 4–6 years3 and substantially higher than that previously recognized.4 Therefore, its high prevalence and long-term impact on individuals, families, and society has important public health implications.2 Moreover, autistic triads may result in other behavioral problems, influence parenting, and probably also have an impact on their siblings.5

In addition to the typical autistic features, children with autism have emotional/behavioral symptoms such as thought problems,6,7 social problems,8 inattention/hyperactivity,7 disruptive behaviors,9 and maladaptive behaviors.10 The Child Behavioral Checklist (CBCL/4-18) has been widely used to measure the behavioral problems,6,7 to validate measures of social behaviors,8,10 or to distinguish the children with autism from those without.6,7 Due to language/communication impairments, however, they may not be able to communicate about their somatic problems, resulting in a lower rate of somatic complaints than normal children.7

Although the literature has documented significant impact of living with or raising a child with autism on parental adjustment and parent–child relationships,11 studies on their siblings have yielded mixed results.12,13 The siblings may demonstrate less positive attitude toward their siblings with autism, regard the latter as a burden, or admire the latter.14 The siblings of children with autism have more problems with peers,14 externalizing problems,13 and internalizing problems13,14 such as depression.15 In contrast, several studies did not support these findings.12,16 Likewise, siblings of children with autism do not differ from their counterparts in self-concepts14 or social adjustment.15 Moreover, some studies found that siblings of children with autism have higher social competence,13 more positive self-concept,13 and better social adjustment16 than their counterparts. Despite mixed results, the findings have converged to suggest that siblings of children with autism may be susceptible to some behavioral or relationship problems but they may not have difficulties in self-concept, social competence or social adjustment.

Many studies on the psychosocial aspects of living with a child with autism have shown that these parents reported less involvement than those of normal and mentally retarded children;17,18 and that fathers are involved less in child care than mothers.19 Konstantareas and Homatidis indicated that the impoverished facial expression, deficits in verbal and gesture communication of children with autism may decrease parent–child interaction.17

Although Western studies have shown that children with autism apparently manifest behavioral problems and their siblings may be at risk for behavioral problems, there is no such study in the Chinese population. Moreover, little is known about the parenting style among children with autism, and to our knowledge, no study has specifically investigated the parenting style for the siblings. Hence, the present study investigated the behavioral characteristics and parenting styles among autistic children, unaffected siblings, and normally developing children to fill the gap between our knowledge of behavioral problems of autistic children and their siblings in the Chinese population and Western literature; and to provide information on parenting style for autistic children and their siblings. We anticipated that mothers of children with autism may be less likely to be employed and have parenting difficulties, children with autism would have a wide range of emotional/behavioral problems, and their siblings may have behavioral problems as well.

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

Participants and procedure

The sample consisted of 151 families with one child with autism as the case families and 113 control families with normally developing children. The 151 index children with autism were mainly recruited from two medical centers (n = 115, 76.2%). The inclusion criteria of the case group were children who had typical autistic features as diagnosed according to DSM-IV autistic disorder, and who had the Catastrophic Illness Card with the diagnosis of autism issued by the Bureau of National Health Insurance, Department of Health, Taiwan. The control subjects were recruited from similar neighborhoods to the case subjects. Those who had a history of developmental delay were excluded from the control group.

The Research Ethics Committee of Taiwan University Hospital approved this study prior to the sample recruitment. Written informed consent was obtained from the parents after explanation of the purpose and procedure of the study, as well as reassurance of confidentiality. Among the 151 children with autism, 106 had one unaffected sibling, and 45 had more than one sibling. Those siblings whose ages were closer to the index children, if there were more than two children in a family and who were older than 2, were recruited into the sibling group, yielding 134 for this group. The number of children whose mother reported on the CBCL for children 4–18 years old was 142 for the autistic children, 120 for siblings, and 109 for the controls. Mothers and fathers reported on the Chinese Parental Bonding Instrument (PBI) about their parenting styles (autistic children, 149 mothers and 147 fathers; their siblings, 134 mothers and 129 fathers; controls: 113 mothers and 109 fathers).

Measures

Child Behavior Checklist

The CBCL is a parental report concerning children aged 4–18. Eight narrow-band subscales were derived from the 118 emotional and behavioral items including attention problems, anxious/depressed syndrome, aggressive behaviors, delinquent behaviors, social problems, somatic complaints, thought problems, and withdrawal.20 Each item was scored as 0 if not true, 1 if somewhat true or sometimes true, and 2 if very true or often true. The Chinese CBCL, a reliable and valid instrument, has been widely used to measure child and adolescent behavioral syndromes in Taiwanese populations.21 The T-scores for each behavioral syndrome were calculated based on a 4–11-year-old norm established in 1998 in Taiwan.21

Chinese version of the Parental Bonding Instrument

The PBI is a 25-item instrument (item-rated on a 4-point Likert scale from ‘very likely’ to ‘very unlikely’) measuring parenting styles during the child's first 16 years with three principal dimensions.22 A high score on the Care subscale (12 items) reflects affection and warmth, while a low score indicates rejection, or indifference. The Authoritarianism subscale (six items) reflects the degree of parental authoritarian control over a child's behavior; and the Overprotectiveness subscale (seven items) reflects overprotective parenting and denial of the child's psychological autonomy.23 The Chinese PBI is a reliable and valid instrument and has been widely used in Taiwanese studies.24

Brief Symptom Rating Scale

The Brief Symptom Rating Scale (BSRS) is a self-administered measure with 50 items rated from 0 to 4 on the basis of the degree of distress caused over the past week. Each item was scored as 0, not at all; 1, a little bit; 2, moderately; 3, quite a bit; or 4, extremely. The BSRS, modified from the Derogatis Symptom Check List-90-Revised, covers nine dimensions of psychopathology: (1) somatization, (2) obsessive-compulsive, (3) interpersonal sensitivity, (4) depression, (5) anxiety, (6) hostility, (7) phobic-anxiety, (8) paranoid ideation, and (9) additional symptoms including vegetative signs and suicidal ideation.25 We used the General Symptom Index (GSI), a mean score of all BSRS categories, as a covariate in the model to control for parental psychopathology.

Diagnosis of autism

The diagnosis of autistic disorder of the full sample was based on the clinical diagnosis according to the DSM-IV criteria for autistic disorder by board-certified child psychiatrists specialized in clinical service of children with autism and confirmed by the Chinese Version of the Autism Diagnostic Interview-Revised (ADI-R). The ADI-R is a standardized and semi-structured diagnostic interview for individuals aged above 2 years.26 The ADI-R interview is applied to subjects' parents or main caregivers. The coding of some items is converted to scores in diagnostic algorithm: 0 if no evidence of abnormality, 1 if some evidence of abnormality, and 2 if evidence of marked abnormality. The algorithm focuses on three domains, which is based on diagnostic criteria of ICD-10 and DSM-IV, including reciprocal social interaction, communication, and restricted, repetitive and stereotyped patterns of behaviors. The ADI-R was translated into Chinese by a team led by SS Gau and YY Wu and approved by the Western Psychological Services.27 The mean scores of the Qualitative Abnormalities in Reciprocal Social Interaction (24.86 ± 5.16) and Communication (18.00 ± 2.48), and the Restrictive, Repetitive, and Stereotyped Pattern of Behavior (7.29 ± 1.49) of the subsample reached the cut-off for autism diagnosis. All of the autistic children had been noted to have autistic symptoms before 36 months old.

Statistical analysis

SAS 9.1 (SAS Institute, Cary, NC, USA) was used to conduct statistical analyses and the pre-selected alpha was 0.05. The three comparison groups were children with autism, unaffected siblings, and unaffected controls. The descriptive results were displayed as frequency and percentage for categorical variables, and as mean ± SD for continuous variables.

We used a linear multi-level model with random and fixed effects to address the lack of independence of children with autism and their siblings within the same family to compare the mean T-scores of eight behavioral syndromes derived from the CBCL, and the mean scores of the mother's and father's parenting styles among the three comparison groups. We used the Bonferroni method to adjust P in the post-hoc analysis. The linear mixed model was also used to compare father's and mother's parenting styles for the same child. All the analyses were controlled for the children's gender and age, and mother's (and/or father's) BSRS scores, educational levels, and employment status. Cohen's d was used to calculate the effect sizes (standardized difference between the two means) for the three groups. The small, medium, and large effect sizes are defined as Cohen's d≥0.2, ≥0.5, and ≥0.8, respectively.

Multiple logistic regression analysis with backward elimination was used to identify the CBCL subscales that might significantly discriminate the autistic children and the controls. Odds ratios (OR) and their 95% confidence intervals (CI) were also calculated.

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

Demographic characteristics

The present results indicated significantly more boys in the autism group than the other two groups (Table 1). Unaffected siblings were significantly older than the other two groups. Families having a child with autism had a higher number of children in the family than the control families (Table 1). There was also significant difference in birth order, with later birth order for autistic children than the controls (P = 0.001).

Table 1.  Characteristics of parents
VariablesAutism (n = 151)Control (n = 113)Sibling (n = 134)Statistics
Mean ± SD or %Mean ± SD or %Mean ± SD or %
  1. *P < 0.05. **P < 0.01. ***P < 0.001.

Age (years)7.8 ± 2.97.9 ± 2.99.0 ± 3.7F(2,395) = 5.31**
Gender, Male87.461.159.0X2 = 34.10***
No. children2.3 ± 0.62.1 ± 0.6 F(1,261) = 9.30**
Mothers    
Current age (years)37.3 ± 4.637.6 ± 4.5 F(1,261) = 0.47
Age at childbirth30.2 ± 3.829.7 ± 4.0 F(1,261) = 0.95
Education level    
Senior high or lower55.038.7 χ2 = 6.77**
College or higher45.061.3  
Employed30.278.3 χ2 = 57.32***
Fathers    
Current age (years)40.3 ± 5.040.7 ± 4.8 F(1,261) = 0.51
Education level    
Senior high or lower36.631.5 χ2 = 0.70
College or higher63.468.5  
Employed97.898.1 χ2 = 0.02

Regarding educational status of the child participants, 84 (55.6%) were primary school students (special educational class, n = 14; resource education, n = 17; regular education, n = 40; both regular education and resource education, n = 13) and 67 (44.4%) were preschoolers receiving early intervention program for autism. For siblings, 93 (69.4%) were primary school students and one of them also received resource education. For the controls, 59 (61.5%) were primary school students and one of them also received resource education.

Mothers of children with autism were significantly less educated and less likely to be employed than mothers of the controls. There was no difference between the two groups in terms of both parents' current ages, maternal age at childbirth, and fathers' educational levels and employment status. The maternal educational level was moderately correlated with that of the fathers (Spearman correlation = 0.55). All the parents of children with autism and the controls were married and lived together, except that two parent dyads of the controls lived separately and one was divorced.

Behavioral characteristics

Table 2 presents the mean T-score of each behavioral symptom derived from the CBCL for the three comparison groups. We found that children with autism had significantly greater severity in all the eight syndromes than unaffected siblings and the controls, with at least small effect size (d ≥ 0.21). The effect sizes (d) were large for attention problems, social problems, thought problems, and withdrawal as compared to the controls (1.59–1.85) and siblings (1.13–1.63). The effect sizes (d) were medium for anxious/depressed symptoms (0.62), aggressive behaviors (0.54), and delinquent behaviors (0.64) as compared to the controls.

Table 2.  Emotional and behavioral problems
CBCL: T scoreAutistic child (n = 142)Sibling (n = 120)Control (n = 109)F(2,113)Autistic child vs SiblingAutistic child vs ControlSibling vs Control
Mean ± SDMean ± SDMean ± SDCohen's dCohen's dCohen's d
  • *

    P < 0.05,

  • **

    P < 0.01.

  • ***

    P < 0.001.

  • Bonferroni method in post-hoc analysis.

  • CBCL, Child Behavioral Checklist.

Aggressive behavior54.6 ± 9.651.4 ± 8.949.8 ± 8.210.59***0.35**0.54***0.18
Anxious/Depressed54.0 ± 10.351.2 ± 9.248.3 ± 7.712.29**0.28**0.62***0.34
Attention problems64.9 ± 11.448.4 ± 8.747.0 ± 7.6115.28***1.63***1.85***0.17
Delinquent behavior51.2 ± 7.549.6 ± 7.747.0 ± 5.411.30***0.210.64***0.40*
Social problems66.7 ± 12.149.7 ± 8.847.9 ± 8.2167.44***1.62***1.83***0.21
Somatic complaints49.5 ± 7.647.6 ± 6.047.6 ± 5.65.36**0.27*0.290.02
Thought problems62.3 ± 13.348.3 ± 7.446.1 ± 5.5118.88***1.29***1.59***0.34
Withdrawal59.7 ± 11.548.4 ± 8.244.2 ± 6.1119.18***1.13***1.69***0.59**
Internalizing55.1 ± 9.049.4 ± 7.246.7 ± 6.350.81***0.69***1.08***0.40
Externalizing53.8 ± 8.750.9 ± 8.348.9 ± 7.212.71***0.34**0.61***0.25

Unaffected sibling had significantly more anxious/depressive symptoms, delinquent behaviors, thought problems, withdrawal, internalizing and externalizing problems than the controls, at a small effect size (0.25–0.40), except withdrawal symptoms, which had a medium effect size (0.59).

The final fitted model using backward selection indicated that attention problems (OR, 1.47; 95%CI: 1.09–1.97), social problems (OR, 1.67; 95%CI: 1.20–2.31), thought problems (OR, 1.84; 95%CI, 1.06–3.17), and withdrawal (OR, 1.69; 95%CI: 1.17–2.42) significantly distinguished children with autism from normally developing controls.

Parenting styles

We found that children with autism and their siblings obtained less affection/care from their mothers than control children, at a small effect size (−0.36; Table 3). The significance disappeared after further controlling for parental GSI score. Moreover, children with autism received significantly less affection/care from fathers than their siblings and controls. Children with autism were more overprotected by their mothers and fathers than were their siblings and the controls (Table 3). Children with autism were under more authoritarian controls by their mothers and fathers than were their siblings. There was no difference between children with autism and their siblings in affection/care from mothers and fathers; or between siblings and controls in affection/care from fathers and in overprotection/authoritarian controls from mothers and fathers.

Table 3.  Parental attitude towards offspring
 Autistic childSiblingControlFAutistic child vs SiblingAutistic child vs ControlSibling vs Control
Mean ± SDMean ± SDMean ± SDCohen's dCohen's dCohen's d
  • *

    P < 0.05.

  • **

    P < 0.01.

  • ***

    P < 0.001.

  • Bonferroni method in post-hoc analysis.

Mother(N = 149)(N = 134)(N = 113)df = (2,133)   
 Affection27.3 ± 4.427.1 ± 5.428.9 ± 4.54.82*0.04−0.36*−0.36**
 Overprotection9.5 ± 3.57.1 ± 3.07.7 ± 3.345.31***0.75***0.56***−0.17
 Authoritative controlling7.5 ± 2.76.2 ± 2.86.7 ± 2.414.65***0.49***0.32*−0.20
Father(N = 147)(N = 129)(N = 109)df = (2,125)   
 Affection24.2 ± 4.525.3 ± 5.126.1 ± 4.86.01*−0.23−0.42**−0.17
 Overprotection9.4 ± 3.17.3 ± 3.27.8 ± 3.128.34***0.66***0.52***−0.15
 Authoritative controlling7.5 ± 2.76.4 ± 2.66.8 ± 2.411.40**0.42***0.26−0.17

Mothers were more likely than fathers to be affectionate to their children for the autism (F(1,144) = 50.96, P < 0.001), sibling (F(1,128) = 12.16, P < 0.001), and control groups (F(1,108) = 20.48, P < 0.001). There was no parental difference in overprotection (P = 0.543∼0.790) and authoritarian control (P = 0.425∼0.842) among the three comparison groups.

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

As the first study to investigate behavioral problems and parenting style among children with autism and their siblings as compared to normally developing children in an ethnic Chinese population, the present study lends evidence to support the hypothesis that more mothers of children with autism were unemployed, children with autism demonstrate more severe behavioral problems in general, their siblings displayed some behavioral problems, parental interaction with autistic children was in a less affectionate/caring and more overprotective/controlling way, and the siblings of autistic children may receive less affection/care and less overprotection/control from their mothers. The unique findings from the present study are that less affectionate interaction with autistic children was more prominent in fathers than mothers, and the siblings also have less affectionate interaction from their mothers, which may be explained by maternal psychological distress.

Behavioral problems of children with autism

The present findings lend evidence to support that children with autism have an increased likelihood of behavioral symptoms in addition to their autistic manifestations. Consistent with previous studies, children with autism in the present study had more severe thought problems,6,7 social problems,8,10 attention/hyperactivity problems,7 and disruptive behaviors.9 Moreover, these children also had more anxious/depressed problems, withdrawn behaviors and somatic symptoms. As expected, the thought problems, withdrawn, social problems, and attention problems had a large effect size and are the most highly associated behavioral problems for autism in the final model. The increased severity of a wide range of behavioral problems may partially support the fact that the CBCL can be used to distinguish children with autism from children without autism.6,7 Inconsistent with the Bolte et al. findings,7 the present study identified a higher rate of somatic complaints among children with autism. Despite communication difficulties, Taiwanese mothers still carefully observed the somatic problems of their autistic children. Because these children tend to have medical conditions,28 caregivers should notice any evidence of physical problems by careful observation.

Behavioral problems of the siblings of children with autism

Although previous studies on the behavioral problems in the siblings of children with autism have demonstrated mixed findings, the present findings suggest that unaffected siblings tend to have anxious/depressive symptoms, a notion supporting previous studies that siblings are more depressive,15 and have withdrawn symptoms, similar to reports of increased feelings of loneliness.14 In addition, the present findings also showed increases in some other emotional or behavioral problems such as delinquent behaviors, thought problems, and internalizing problems among siblings of autistic children. The present study did not provide evidence to support increased severity of a wide range of behavioral problems among the siblings of autistic children with the exception that the siblings may have more severe delinquent and withdrawal symptoms than normally developing children with small to medium effect sizes. Because these siblings were not clinically diagnosed as having ASD, the effect of autistic features on the behavioral problems can be ruled out, but there is no information about the existence of autistic trait in the siblings that may not meet the diagnostic criteria for ASD. Unlike others,12,16 the present findings suggest that it is important to conduct early screening for behavioral problems and autistic trait in the siblings of children with autism.

Parental characteristics and parenting style

In the context of Chinese culture, families in Taiwan are greatly influenced by the Confucian tradition, which emphasizes patriarchy in the family, the family over the individual, children's academic achievement, and hierarchical relationship and responsibility.29 Mothers of children with autism may take even more responsibility for caring and educating a disabled child and the fathers may be less likely to accept having a child with autism because of the emphasis on patriarchy and family honor in Chinese families. Consistent with Western studies,30 the present findings showed that mothers of children with autism were more likely than their counterparts to take care of their children full-time rather than being employed. The reduced maternal employment due to the increased parental care needs of children with autism or other disabilities is a cross-culturally common phenomenon rather than being unique to the Chinese culture.

Although the increased stress and demands on the parents of children with autism or other disabilities have been documented in the literature,11,30 relatively little research has specifically investigated the parenting style among children with autism and their siblings, particularly, to our knowledge, no study has been conducted in the Chinese population. In contrast to our prior finding of similar parenting style between children with Down syndrome and normal children,31 the present findings suggest less affectionate, more protective, and more controlling parenting in children with autism than the other two groups. The abnormal reciprocal relationship, lack of or decreased response to social stimulations, and communication impairments in children with autism may decrease parents' affectionate interactions with these children;17 and moreover, these children's typical autistic features may result in increased needs of care and heightened parental overprotection and controls. Therefore, disability or developmental retardation itself may not be a predictor for affectionless, overprotective, or authoritarian parent–child interaction, which may be predicted by autistic features. The discrepancy in findings between the previous study in children with Down syndrome31 and children with autism lends some evidence to support prior findings that parents of autistic children reported less interaction,18 involvement and contact17 than parents of children with Down syndrome, with mental retardation, or without disabilities. This assumption, however should be tested in future studies by comparing the parenting styles among children with different kinds of disabilities.

Additionally, we found that mothers of children with autism also had decreased affectionate interaction with their children without autism. As expected, fathers were less affectionate toward their children than mothers. In contrast to the fact that mothers of children with autism treated their children equally less affectionately than mothers of controls, fathers of children with autism were significantly less affectionate toward their children with autism than their non-autistic children, a finding similar to Western studies.17,19 Moreover, as in the Konstantareas and Homatidis study,17 the discrepancy of fathering and mothering was greatest in the autism group, suggesting that having a child with autism may influence fathering more than mothering. The question of whether there is a cross-cultural difference in the extent of the differential parenting with regard to having a child with autism and between mothering and fathering, needs future cross-cultural studies to provide evidence.

Limitations

The present study was one of few studies examining the parenting style among children with autism and their siblings, particularly in non-Western countries. The other strengths of the present study included the large sample size, and inclusion of sibling controls and assessment of paternal parenting styles by fathers' reports.

Despite these strengths, methodological limitations should be considered when interpreting the findings. Restriction of the autism group to those children who were diagnosed with autistic disorder and had received special education before 6 years of age has minimized the heterogeneity of this group, which may decrease the generalization of the present study. Although the siblings were assessed clinically to rule out the diagnosis of ASD, lack of information on autistic traits that did not fulfill the criteria for a diagnosis of ASD during the developmental stage has precluded our ability to examine the effect of autistic trait on behavioral problems and inappropriate parenting style among the siblings.

Implications

The present findings imply that in addition to emphasis on the behavioral problems and psychosocial care of children with autism, it is necessary to conduct early screening for the possibility of autistic trait and other behavioral problems for siblings of children with autism. Moreover, fathers had a greater degree of significantly decreased parental affection and care toward their children with autism than mothers did. Therefore, parenting counseling for this population should focus not only on children with autism, but their sibling as well; and should involve not only mothers as is traditional in Taiwan, but fathers as well.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES

This work was supported by a grant from the National Science Council (NSC 93-3112-H-002-005) and the preparation of this manuscript was supported by the National Health Research Institute (NHRI-EX94-9407PC), Taiwan. The authors would like to express their thanks to the parents for their contribution to the present study.

REFERENCES

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. REFERENCES