- Top of page
Summary: Purpose: To examine the contribution of epilepsy-related factors, proximal (parent–child relationship quality), distal (parental characteristics), and contextual (quality of other family relationships) family factors to psychopathology (both broad-band and narrow-band syndromes) in children with epilepsy and normal intelligence.
Methods: Parents of 91 children (mean, 8.5 years) participated by filling out questionnaires about family factors and child psychopathology. Regression analyses were used to analyze the unique and combined predictive power of family factors in relation to psychopathology.
Results: In contrast to epilepsy-related factors, family factors, especially those related to the quality of the parent–child relationship, appeared to be strong predictors of psychopathology. The results supported the mediational model: Distal and contextual factors affect child psychopathology by affecting proximal factors.
Conclusions: In treating children with epilepsy, clinicians should be aware of the importance of the parent–child relationship quality. Strengthening the relationship quality may prevent or reduce psychopathology.
The literature provides ample evidence that psychopathology is more common in children with epilepsy than in children from the general population (1,2). A recently conducted meta-analysis confirmed that children with epilepsy have higher levels of psychopathology than do children from the general population (3).
The factors that have been found to contribute to psychopathology in children with epilepsy are multiple (4,5). They consist of neurologic factors (e.g., age at onset, epilepsy duration, epilepsy severity, seizure type, and seizure control) (6,7), medication factors such as side effects (8), and psychosocial factors (e.g., perceived stigma, attitude toward illness, and family factors) (9–13). While reviewing the behavioral and cognitive correlates of epilepsy in 1984, Hermann and Whitman (5) pleaded at that time for a more serious examination of family factors as determinants of psychopathology in children with epilepsy. Since then, the examination of distinct family factors as contributors to child psychopathology has gradually been incorporated in childhood epilepsy research.
Family-factor constructs represent different aspects of family functioning, but to date, an appropriate model to differentiate between family factors has frequently been lacking in childhood epilepsy research. It is essential, though, to categorize family factors into distinct types, because each type of factor is considered to play a different role in relation to child outcome (14–16). Following social interactional and ecologic perspectives, family factors can be ordered, according to the level of proximity to the child's everyday life, into proximal family factors (the quality of the parent–child relationship and parenting), distal family factors (parental characteristics), and contextual family factors (the quality of other family relationships) (14,17–19).
Although it can be deduced from childhood epilepsy literature that distinct family factors are associated with psychopathology in children with epilepsy (20), the simultaneous examination of distinct family factors seldom occurred. Little is known about the unique and the combined predictive power of each of these family factors as contributors to child psychopathology. The examination of the interrelationships among distinct family factors themselves and the pathways between these distinct family factors and child psychopathology may add new information, which may be helpful for children with epilepsy and their families.
With the current study, we aim to examine the contribution of epilepsy-related factors and distinct family factors—ordered into proximal, distal, and contextual factors—to psychopathology in children with epilepsy. Even though it is possible to make clear conceptual distinctions between family factors, it also is clear that factors representing family functioning do not function separately from each other within families, but are interrelated constructs that in joint interaction exert their influence on the child's everyday life (21). As a consequence, it may be that spurious associations arise between distinct family factors and child psychopathology. Therefore with this study, we aimed to examine the independent contributions of proximal, distal, and contextual family factors to child psychopathology and to examine whether certain types of family factors predominate in predicting child psychopathology (18). Both family risk factors and protective factors were selected, because it is considered that child outcome is affected by the equilibrium between family stressors and protective (buffering) factors. That is, according to the cumulative risk model, the chance that psychopathology develops in children increases when family risk factors outweigh protective family factors (15,16,22).
Although childhood chronic illness literature has emphasized the importance of applying a noncategoric approach to examining the effects of chronic conditions on child development (that is, generic features of chronic conditions such as functional status, the limitation of activities, or the burden of daily care on the family) (23,24), disease-specific parameters also have been consistently depicted in conceptual models that explain child adjustment in chronic conditions (19,25). Among these disease parameters are condition severity, which has been found to be a contributor to child adjustment (26) and condition duration, which received less attention in childhood chronic illness research (25).
In childhood epilepsy, disease parameters could be described as pathogenetic contributors of psychopathology, that is, central nervous system (CNS)-related or epilepsy-related factors (27). When epilepsy-related factors were examined with parenting factors, it was found that parenting factors predicted internalizing and externalizing behavior problems better than the effects of epilepsy-related factors (12,13). However, significant relations also have been found between seizure severity and child psychopathology (28), including internalizing, externalizing, attention, and thought problems (7,29). Hoare and Mann (30) found an association between longer epilepsy duration and child psychopathology, including depression and anxiety (31).
Proximal family factors comprise the quality of the parent–child relationship and parenting (18,32). Parent–child relationship quality is defined as a constellation of parental attitudes that has been built up in the long history between parent and child, in which the parent's behaviors are expressed (18,32). Good quality in terms of the parent–child relationship has generally been associated with lower levels of child psychopathology in samples of both healthy children and children with epilepsy (9,13,33–35).
In contrast, low parent-child relationship quality, such as parental rejection, has been found to exert negative effects on child development and is thus considered to be a risk factor for the development of child psychopathology (36–38). One study that measured expressed emotion (EE) in mothers of children with epilepsy showed that high maternal criticism, an indicator of a low parent–child relationship quality, was related to higher levels of child psychiatric disturbances and antisocial behavior (33).
Distal family factors are dispositional parental characteristics that are assumed to influence the child indirectly, by disrupting the quality of parenting (18). Maternal depression is a distal family factor that is of current interest in child developmental and family research, and also in childhood epilepsy literature. Child developmental literature consistently showed both direct associations between maternal depression and child psychopathology (39,40) and indirect associations between maternal depression and child psychopathology by disrupting the quality of parenting (18,41). Childhood epilepsy literature showed that ∼30% of the mothers of children with epilepsy are at increased risk for depression and showed associations between maternal depression and higher levels of child psychopathology (33,42,43).
Counter to the risk-factor depression, we selected another dispositional characteristic that protects against psychopathology, parental competence in parenting. Although this characteristic has scarcely been researched in relation to child psychopathology in the epilepsy literature (9), in the child developmental literature, it has frequently been linked to adequate parenting (44,45), which in turn is related to lower levels of child psychopathology (46,47).
Contextual family factors are often understood as indicative of psychopathology in individual family members, reflecting overall family dysfunction (48). In light of childhood chronic conditions, family adaptation is commonly seen as a key factor for healthy child adjustment (49,50). Childhood chronic conditions are considered to affect the whole family, requiring new modes of organization and structure for the family (51). The family failure to adapt adequately to the demands of childhood chronic illness could, therefore, be considered a risk factor for development of psychopathology. Studies with samples of both healthy children and children with a chronic illness, including children with epilepsy, showed a relation between problems with family adaptation and the development or maintenance of child psychopathology (9,52–54).
Marital satisfaction also has frequently been linked to child psychopathology. Unsatisfactory marital relationships have been shown to function as a risk factor for the development of child psychopathology by interfering with parent–child relationship quality (55–57). In contrast, good marital relationship quality is positively associated with adaptive child outcome (58). To date, marital relationship quality has been scarcely researched in samples of children with epilepsy. One study, carried out by Austin, Risinger, and Beckett (59), found that marital strain was correlated with higher levels of child behavior problems. Another earlier study (60) found that a disrupted parental marriage contributed to internalizing and externalizing behavior problems in children with epilepsy.
The relation between family factors and psychopathology in children with epilepsy has been examined mainly on the level of Internalizing (e.g., Anxiety/Depression and Withdrawal) and Externalizing (e.g., Aggression and Delinquency) broad-band behavior problems, measured with the Child Behavior Checklist (61,62). The relation between the syndrome scales (Anxiety/Depression, Withdrawal, Somatic complaints, Delinquency, Aggression, Social problems, Attention problems, and Thought problems) and family factors has been less frequently assessed in children with epilepsy (20). To overcome this gap and, in particular, to study thought problems and attention problems that were considered relatively specific to epilepsy (3), the current study includes all the syndrome scales.
In sum, we aimed to examine simultaneously the relation between epilepsy-related factors, distinct family factors, and distinct types of psychopathology in children with epilepsy by examining the independent contribution of each cluster of epilepsy and family factors to child psychopathology. More specifically, concerning the relation between distinct family factors and child psychopathology, two mechanisms may play a role. First, it may be that each family factor has an independent effect on child psychopathology. Although we expect that each of these factors would be related to child psychopathology, we expect that the proximal factors would exert the greatest influence on child psychopathology (16). Second, it may be that family factors are interdependent in their influence on child psychopathology. On basis of the social interactional and ecologic model, it is subsequently assumed that the effects of distal (parental depression and parental competence) and contextual factors (problems with family adaptation and marital satisfaction) are mediated by proximal family factors. In other words, the distal and contextual family factors are expected to affect the parent–child relationship (proximal factors), which, in turn, would affect child psychopathology.
- Top of page
The goal of this study was to explore the relative influence of epilepsy-related factors and distinct family factors on child psychopathology. Based on the social interaction and ecologic model, we hypothesized that the most proximal interactions between parent and child (parent–child relationship quality) would exert the greatest influence on child psychopathology. In addition, it was hypothesized that proximal family factors would mediate the effects of distal and contextual family factors on child psychopathology.
First, however, the severity of child psychopathology in our sample was compared with the norm group. The comparison showed that children with epilepsy had significantly higher levels of psychopathology. This holds particularly for the presence of Internalizing problems, Externalizing problems, and Attention problems, thereby confirming findings from the existing childhood epilepsy literature that indicated that children with epilepsy are at higher risk for the development of psychopathology (1,2,83).
Considering the relative influences of epilepsy-related factors and distinct family factors on both broad-band and narrow-band behavior problems, it appears that epilepsy-related factors were nonsignificant contributors, irrespective of the type of psychopathology, whereas each type of family factor significantly influenced almost each type of child psychopathology (an exception appeared for Somatic complaints, distal factors as predictors of Anxiety/Depression, and contextual factors as predictors of Internalizing behavior problems). However, when each type of family factor that appeared to be significant in the first step of the regression analyses was entered in the last step (i.e., when the other significant family factors were controlled for), only proximal family factors exerted influence on child psychopathology, whereas former significant effects of distal and contextual factors disappeared. These findings support the social interaction and ecologic model in that the effects of the most proximal family factors on child psychopathology are stronger than the effects of the more distal and contextual family factors (15–17). Internalizing behavior problems (including Depression and Withdrawal) and Thought and Attention problems were solely predicted by three family risk factors: rejection, depression, and family adaptation problems. In contrast, a better equilibrium appears to exist between stressors and compensatory factors in case of Externalizing behavior problems (at least as many protective factors as risk factors are found).
Besides that the parent–child relationship quality was revealed to be the most important contributor to child psychopathology, even when the other family factors were controlled for, the results also confirm our hypothesis that proximal family factors would mediate the effects of distal and contextual family factors. First, parental depression and marital satisfaction influence Externalizing behavior problems and Aggression through parent–child relationship quality (both rejection and positive parent–child relationship quality). Second, the effects of parental depression and marital satisfaction on Delinquency are mediated by the positive parent–child relationship quality. Third, parental depression and problems of family adaptation (except for Internalizing behavior problems) exert their influence on Internalizing behavior problems, Anxiety/Depression, Withdrawal, and Thought and Attention problems through parental rejection. Especially in the latter case, the adverse effects of distal and contextual family risk factors seem to spill over into poor parent–child relationship quality, which in turn, negatively affects child psychopathology.
It should be pointed out that, because of the cross-sectional nature of this study, we are not able to draw any conclusions regarding causality. Although we assume that family factors contribute to the development of psychopathology in children with epilepsy, it is just as possible that problems within the family are reactions to already existing pathology and thus emerge as contributors to the maintenance of psychopathology. Besides, it may also be true that child psychopathology contributes to higher levels of maternal depression, family adaptation problems, or marital conflict, which in turn, contribute to the maintenance of child psychopathology. Such pathways may be addressed in future research that examines pediatric epilepsy from a child-developmental psychopathology perspective (84).
Regardless of the direction of the effects, the results clearly show that family factors are more strongly related to psychopathology in children with epilepsy than are epilepsy-related factors. This may indicate that it is especially the presence of a chronic condition (i.e., generic factors) that places demands on the family, rather than the effects of specific disease parameters (e.g., severity and duration). It may be that the burden of epilepsy interacts with family processes and that these factors form together a complex web of contributors to child psychopathology (19,51). Therefore it may be of importance for future childhood epilepsy research to use generic effects of epilepsy (for instance, the degree to which epilepsy affects the child's functional status, visibility and stability of the condition, the need for extra medical care, and the impairments of daily activities) (19,85). In addition, it might be important to study parental beliefs and child beliefs about the child's illness and treatment (10,11,13).
From meta-analytic results, it was recently deduced that family factors were less strongly involved in Thought and Attention problems, as these were found to be relatively specific to epilepsy (3). This study showed, however, that distinct family factors also are strongly involved with Thought and Attention problems. Therefore the contribution of family factors to Thought and Attention problems may be attributed to reactions of the family to already existing pathogenetic causes of psychopathology.
Moreover, as many pathways explain psychopathology in children with epilepsy (16,86), it may be that the influence of pathogenetic contributors is easier to detect in other childhood epilepsy populations. For example, it might be that symptomatic epilepsy syndromes, which involve central nervous system (CNS) lesions, are, among other factors, strong contributors to child psychopathology (27). Therefore future research should focus on longitudinal studies with large samples, which would enable us to examine transactional pathways of social ecologic contributors and pathogenetic epilepsy contributors to child psychopathology (19,87).
Some other limitations also should be addressed. First, psychopathology was assessed with the CBCL, a screening instrument that examines the risk at psychopathology. Although it has consistently been demonstrated that the empirically based CBCL scales converge with the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) diagnostic categories (88,89) use of the CBCL in clinical samples should be considered with some caution. For instance, Perrin et al. (85) stated that the interpretation of Somatic complaints in clinical populations should be done cautiously, because it is difficult to discriminate between symptoms that may reflect psychopathology and symptoms that reflect the effects of chronic conditions or reflect side effects of medical treatment itself. Besides, Perrin et al. also stated that Somatic complaints could be reflections of psychopathology but nevertheless are interpreted as consequences of a chronic condition. Moreover, it has been suggested that certain behavioral items on the CBCL checklist reflect seizure features rather than behavior problems (90). However, if parents of children with epilepsy were asked to exclude behaviors that they considered to be seizure specific, parents still rated their children higher on levels of psychopathology than did parents of children from the general population (6).
Second, psychopathology and distinct family factors were measured by means of parental questionnaires only. Therefore the findings of the present study should be ascribed to parent's reflections on their own behaviors, beliefs, and attitudes as well as to reflections on their children's behaviors and should be interpreted with caution. In particular, solely using parental report of child psychopathology may lead to shared method variance. That is, parental depression may distort parental ratings of child psychopathology, leading to artificial stronger relations. For instance, maternal reports of internalizing behavior problems, but not externalizing behavior problems, were significantly biased by maternal psychopathology (91). With regard to the current study, shared method variance should then be ascribed to the more internalizing types of child psychopathology rather than to the more externalizing and unambiguous types of psychopathology such as attention problems and thought problems. It also has been demonstrated that, contrary to former research, depressed mothers are accurate reporters of their children's behaviors (92,93), indicating that parental depression does not distort parental ratings of their children's behavior.
In addition, although inconsistent findings exist about the convergence between parental reports of parenting and other informants of parenting (e.g., 94,95), it has been found that mothers are not inclined to rate their parenting behaviors more positively (95). Because parental subjective experiences are of importance for child behavior and development, it can then be considered that, in particular, in the context of parenting, parents seem to be optimal informants of rating family factors and child behavior. Therefore we concur with the optimal informant strategy (96) that assumes that the person most engaged in a particular context is also the person in best position to rate the constructs belonging to that context. This strategy, however, may eliminate additional, unique information of other informants (e.g., child report). Therefore multisource (by means of multimethod) reporting of psychopathology and social ecologic factors may elucidate how parental, child, and environmental influences affect psychopathology in children with epilepsy, which, as a consequence, should be more consistently applied in future research (96).
Third, as children with idiopathic epilepsy syndromes constituted the majority of children in this study, the research findings may not be representative for the whole population of children with epilepsy. Seizure-condition severity is generally assumed to contribute to adjustment in childhood epilepsy, but one of the contributors to condition severity—epileptic syndrome severity—is seldom included in research. Dunn et al. (97) recently pleaded for the inclusion of epilepsy-syndrome severity in the measurement of condition severity, and they developed a syndrome-severity rating. Therefore future childhood epilepsy research should examine whether certain samples of children (children with symptomatic epilepsy) are at higher risk for psychopathology and whether the pathways that explain psychopathology are different for these particular groups of children.
This study showed that when epilepsy-related factors and distinct family factors are examined within a social interaction and ecologic framework, it is the most proximal family factors that exert the greatest influence on psychopathology in children with epilepsy. In particular, parental rejection appeared to be a risk factor that contributes to Internalizing behavior problems, Withdrawal, Depression, Thought, and Attention problems. Moreover, with respect to the same types of psychopathology, rejection appeared to be a mediator of two other family risk factors: depression and problems with family adaptation. Thus clinicians should be aware of the detrimental effects of negative parent–child relationship quality on child psychopathology and the risk factors that are associated with parent–child relationship quality. In contrast, positive parent–child relationship quality may compensate for adverse effects of family risk factors in case of Externalizing behavior problems. Therefore in treating children with epilepsy, attention should be paid to the quality of the parent–child relationship and to the use of strategies that ameliorate and strengthen it.