- Top of page
Aim: To investigate the effects of age, gender, prenatal/perinatal factors, and maternal psychological distress on childhood emotional/behavioral problems, and the gender difference in the patterns of comorbid emotional/behavioral problems.
Methods: The sample included 1391 children aged 4–9 in Taipei using a multi-stage sampling method. Their mothers completed questionnaires including demographics, prenatal/perinatal variables, the Child Behavior Checklist, and the Chinese Health Questionnaire. The linear and non-linear mixed model was used for data analysis.
Results: Boys scored higher on externalizing problems, and girls scored higher on internalizing problems. Gender also modified the comorbid patterns of emotional/behavioral problems. Aggressive behaviors decreased, but attention and thought problems increased with age. Prenatal/perinatal exposure to alcohol and coffee, vaginal bleeding, and gestational diabetes, low birthweight, and postnatal incubation and resuscitation, and maternal psychological distress predicted the risk for several childhood emotional/behavioral problems.
Conclusion: Prenatal/perinatal and maternal care, and gender-specific measures are important for prevention of childhood emotional/behavioral problems.
CHILDHOOD BEHAVIORAL PROBLEMS are common and related to impairments in self-competence and family function.1 Early behavioral problems are predictive of poor adjustment and psychopathology in late adolescence.2 Therefore, it is essential to identify correlates and risk factors of childhood behavioral problems for the delivery of mental health services.
Literature has documented more externalizing problems in boys, and more internalizing problems in girls;3,4 and increased somatic complaints and withdrawal with age for boys, but decreased for girls.5
There is accumulating evidence that prenatal/perinatal exposure may have long-lasting adverse consequences on the brain development and increased behavioral problems of the offspring. For example, low birthweight predicts inattention/hyperactivity symptoms and clinically significant behavioral problems over time during childhood.6,7 In addition, prenatal exposure to maternal substance use such as nicotine and illicit drugs may be linked causally to childhood behavioral problems in a positive dose–response relationship.7,8 These findings, however, are not supported by others.9
Several studies have suggested an association between parental mental distress and children's emotional/behavioral problems.10–12 For example, the most powerful predictor of high scores on the Child Behavioral Checklist (CBCL) among premature preschoolers is maternal self-reported depressive symptoms.11,12 Although previous work has shown that maternal psychological distress is associated with several types of sleep problems in children,13 no study has been done to examine its relationship with children's emotional/behavioral problems.
Similar to the findings in adolescent and adult populations, several studies also report a high comorbidity of behavioral problems in children,14,15 and gender differences in comorbid patterns.15 Outcome of comorbid behavioral problems is worse than that of the components alone.16 These findings highlight the importance of identifying comorbid behavioral patterns in childhood to offset future development of psychopathology and adverse social outcome.16
Despite several epidemiological studies on Taiwanese child and adolescent populations,13,17–19 there has been no specific study to investigate the age and gender difference in the comorbid behavioral problems patterns, or to examine the effect of prenatal/perinatal factors on behavioral problems in young children of an ethnic Chinese population. Hence, we conducted an epidemiological survey of childhood behavioral problems in Taiwan to investigate the effects of age, gender, prenatal/perinatal factors, and maternal psychological distress on childhood emotional/behavioral problems, and the gender difference in the patterns of comorbid emotional/behavioral problems. We hypothesized that there would be age trends and gender differences in the rates and comorbid patterns of childhood emotional/behavioral problems, and that childhood emotional/behavioral problems would be associated with some prenatal and perinatal exposure, and maternal psychological distress.
- Top of page
The present findings lend evidence to support the effect of gender and developmental stage on the manifestation of emotional/behavioral problems. The major findings are that there was an association between childhood behavioral problems and a multiplicity of risk factors, including exposure to alcohol and coffee, antepartum vaginal bleeding, gestational diabetes, low birthweight, postnatal incubation, and postnatal resuscitation and maternal psychological distress. Moreover, the comorbid patterns of emotional/behavioral problems in boys did not differ much from those in girls.
As the first study examining the childhood behavioral problems in Taiwan, the figures of symptom severity of the eight behavioral symptoms measured on the CBCL were more severe than those found in the children aged 6–11 years in China.4 This discrepancy might be explained by the different cultural, educational, and societal context between Taiwan and China despite the similar ethnicity.
Consistent with Western studies, the present findings suggest that boys have more externalizing problems (i.e. aggressive and delinquent behaviors) and girls have more internalizing behaviors (i.e. anxiety/depression symptoms and somatic complaints).3,23 Moreover, the present findings also support the notion that inattention/hyperactivity and thought problems are more prevalent in boys than girls.18,24 Despite the consistent findings across studies, little has been known about the underlying mechanism of the gender differences in childhood behavioral problems.
Similar to several studies, the present findings add evidence to support a reduction in aggressive behaviors, indicating the possible relationship with psychosocial maturation from preschool to middle childhood.3–5 The negative finding, however, of significant decrease in delinquent behaviors or social problems with age3 may be explained by the generally low scores of these two syndromes in early and middle childhood in the present sample. Although previous studies on the age trend of attention problems produced inconsistent results,5,25 the present findings indicate that attention problems increased from early to middle childhood. It is highly likely that Chinese parents reported more severe child attentional problems with age due to the increased school load and high expectations of school performance despite the developmental maturation of attention.25 The novel finding is increased score on thought problems with age, which may be attributable to the increased verbal expression and cognitive development with age with which to report their imagination and fantasy.
Comorbid patterns of behavioral problems
Overall, the rates of comorbid conditions of particular behavioral problem pairs in the present study ranged from 0.9% to 4.0%, lower than those reported in the USA (10.5–30.2% in children aged 4–18).14 This discrepancy can be explained by different definitions of comorbidity and the age range of the participants. This presents a challenging issue for the reliable assessment of comorbidity rates of childhood behavioral problems.15 The present findings demonstrated similar comorbid patterns in childhood to Western studies.14 Gender difference in some comorbid behavioral patterns is a unique finding, which has been relatively less studied with mixed findings.15
In general the present findings provide some evidence of prediction of some prenatal and perinatal factors to childhood emotional/behavioral problems. Consistent with other studies,8,26 the present findings suggest that exposure to alcohol increases the risks for some childhood behavioral problems. Despite the mixed results27 the present findings demonstrate that prenatal exposure to regular coffee use increased social problems in children. Caffeine and ethanol, which can cross the placental barrier and reach the human fetal brain,28 should be investigated as to what extent prenatal exposure to these common substances contributes to the development of child behavioral problems. Unlike other studies,7 the present study does not lend support to the impact of maternal smoking during pregnancy.
Consistent with a previous study, the association between antepartum vaginal bleeding and gestational diabetes, and behavioral problems implies that fetal adaptation to an unfavorable intrauterine environment may increase susceptibility to childhood behavioral problems.29 In addition, the present findings and several lines of data support an association of low birthweight with inattention/hyperactivity in childhood.6 Moreover, the findings of the impact of postnatal resuscitation and incubation on attention problems and social problems suggest a possible consequence of neurodevelopmental disability after neonatal resuscitation.
Overall, the present findings lend some support to increased rates of a variety of behavioral problems in children born with prenatal and perinatal insults, which may not necessarily be causal in nature but have implications for prevention of childhood behavioral problems.4
Maternal psychological distress
The present study demonstrates an association between maternal psychological distress and children's emotional and behavioral problems.7,11 Parental psychological distress may modify the parent–child relationship at a number of levels and in a reciprocal manner to increase the risk of behavioral problems.7 The effect of maternal depression on childhood emotional and behavioral problems has been studied the most;10 this relationship was further supported by a reduction in children's current behavioral symptoms after maternal remission from depression.10 It would be informative to examine whether intervention for maternal psychological stress ameliorates children's behavioral problems, in a future experimental study.
Several methodological limitations should be considered when interpreting the present findings. First, the mother was the only informant for reporting on prenatal and perinatal exposure and the child emotional/behavioral problems. Neither medical records nor a second informant was used to validate these data. Studies have shown, however, that the mother is the most appropriate informant to report on these variables;30 recall bias, differential or non-differential, is minimized due to the young age of the child subjects, the mothers were not aware of the study hypothesis. Second, lack of quantity assessment of prenatal substance exposure prevents us from examining the dose–response relationship between prenatal exposure and childhood behavioral problems. Third, although maternal reports on children's emotional/behavioral problems are prone to observer bias and are influenced by subjective expectation and issues related to parent–child interaction, parents' reports on children's behaviors have been documented to be the satisfactory way to measure behavioral problems in children.30 Questionnaires and rating scales with good psychometric properties (e.g. the CBCL) are valuable in the evaluation of children with different kinds of behavioral problems.5 Furthermore, because a home-based interview approach would likely have yielded a lower attrition rate,23 the questionnaire survey method is alternatively satisfactory and cost-effective in achieving an adequate response rate, and interviewer bias can be prevented.
The strengths of the present study were the recruitment of a large-scale epidemiological sample using the multistage sampling method, with a satisfactory response rate and use of reliable and valid measures.