- Top of page
Summary: Purpose: Based on prior research indicating poor health-related quality of life (HRQOL) in children with attention-deficit/hyperactivity disorder, we investigated (1) whether executive functioning deficits were related to poor HRQOL in children with epilepsy, (2) how important these variables were in comparison to known predictors of HRQOL such as neurological factors, and (3) the extent to which clinical-level impairments in executive dysfunction predispose children to low HRQOL.
Method: Data included scores on the Behavior Rating Inventory of Executive Function (BRIEF) and HRQOL scales (The Impact of Childhood Illness Scale [ICI] and Hague Restrictions in Epilepsy Scale [HARCES]) for 121 children (mean age = 11.9, SD = 3.6) from a tertiary center serving children with severe epilepsy.
Results: Correlations between the BRIEF and ICI total and subscore domains (child, parent, family, and treatment) were generally significant and moderate (e.g., r ≥ 0.30, p ≤ 0.001). BRIEF Global Executive Composite, number of antiepileptic drugs (AEDs), number of prior AEDs, and adaptive level all emerged as significant and unique predictors of HRQOL (R2= 0.36, adj. R2= 0.33, p < 0.0001). A clinically elevated BRIEF was associated with a twofold risk of low HRQOL (odds ratio = 2.21, p = 0.03).
Conclusions: Executive dysfunction appears to exert a broad adverse influence on HRQOL in children with epilepsy, with clinical-level impairments in executive dysfunction contributing to a twofold increase in the likelihood of poor HRQOL. The constellation of executive dysfunction, low adaptive level, high medication load, and a history of several failed AEDs are risk factors for poor HRQOL in children with epilepsy.
Executive functioning refers to critical functions such as planning, inhibition, set shifting, self-monitoring, organization, working memory, and initiating and sustaining motor and mental activity. Executive dysfunction is associated with behavioral disturbance, social dysfunction, and reduced educational and occupational attainment (Baron, 2004; Lezak et al., 2004). Health-related quality of life (HRQOL) is defined as encompassing the individual's well-being in psychological, social, occupational, and educational domains (Duncan, 1990; Austin et al., 1996). Because these domains are among those adversely impacted by executive dysfunction, it should follow that childhood disorders associated with executive dysfunction might be at increased risk for decreased HRQOL. Emerging research suggests that this is indeed the case: disorders such as attention-deficit/hyperactivity disorder (ADHD) and head injury, both associated with executive dysfunction as core symptoms (Barkley, 1997; Baron, 2004), present the most compelling argument for an association between executive dysfunction and HRQOL. Research involving these disorders indicate an increased risk of poor HRQOL in several domains, including physical functioning, social functioning, parental emotional health and activities, family activities, and family cohesion (Graetz et al., 2001; Klassen et al., 2004; Matza et al., 2004; Topolski et al., 2004; Escobar et al., 2005; Horneman et al., 2005).
Studies examining the impact of executive dysfunction on the HRQOL of children with epilepsy are entirely lacking, despite the ubiquitous nature of executive problems in children with epilepsy presenting for treatment at tertiary centers. In one study, 40–50% of children with epilepsy had clinically significant problems with planning or working memory, key elements of executive functioning (Slick et al., 2006). In another study, almost two-thirds of children with epilepsy were found to have ADHD symptoms (Thome-Souza et al., 2004). If there is a link between executive dysfunction and HRQOL in epilepsy, this would mean that a significant proportion of children are at risk for poor HRQOL by virtue of their executive deficits, in addition to their increased risk of poor HRQOL secondary to medical, psychological, cognitive, and sociodemographic factors. These include factors such as low IQ, low adaptive functioning, psychosocial difficulties, high seizure frequency, intractability, polydrug therapy, longstanding epilepsy duration, parental maladjustment, low family income, and older age (Devinsky et al., 1999; Sabaz et al., 2001; Sherman et al., 2002; Buelow et al., 2003; Miller et al., 2003; Sabaz et al., 2003; Williams et al., 2003; Sillanpaa et al., 2004).
We aimed to determine whether executive functioning deficits were associated with decreased HRQOL in children with severe epilepsy, and if so, whether these were major predictors of HRQOL compared to other known predictors of HRQOL in pediatric epilepsy. The rationale was that any predictor of HRQOL, to have clinical utility, would have to have at least a moderate association with HRQOL, and that this association would remain when other known predictors of pediatric HRQOL such as neurological variables and adaptive level were simultaneously considered.
- Top of page
While we expected executive functioning to be associated with HRQOL based on prior research indicating poor HRQOL in children with ADHD, we were surprised at the extent to which BRIEF scales predicted HRQOL in the context of pediatric epilepsy. Unlike ADHD, pediatric epilepsy is a neurological condition defined by seizures, not behavioral disturbance; further, unlike ADHD, it is a chronic medical condition associated with several known neurological predictors of poor HRQOL that together would be presumed to account for a significant proportion of variance in HRQOL. Executive dysfunction was presumed to be a comparatively weaker predictor given that the central focus of children and families, particularly those struggling with intractability, was presumed to be the seizure disorder itself. This is not what we found; instead, executive dysfunction was an important predictor of HRQOL, equivalent to neurological predictors such as number of AEDs and number of failed AED trials, two markers indicative of the severity and extent of intractability of the seizure disorder. The results suggest that executive dysfunction is a significant barrier to the HRQOL of children and families and that executive dysfunction exerts a broad adverse influence on several components of HRQOL. In this study, clinical-level impairments in executive dysfunction were associated with a twofold increase in the likelihood of low HRQOL. In particular, the constellation of executive dysfunction, low adaptive level, high medication load, and a history of several failed AEDs appeared to contribute significantly to the risk of poor HRQOL in children with epilepsy.
Although most aspects of executive functioning as measured by the BRIEF were related to HRQOL, not all dimensions were associated with HRQOL. Examination of the correlation matrix in Table 4 suggests that at the composite level, an index of self-control/inhibition (BRI) was related to almost all aspects of HRQOL. At the subscale level, the Working Memory scale was also related to most dimensions of HRQOL, particularly with regard to the frequency with which the child's life was adversely affected. Importantly, although clinical elevations on this scale are frequent in pediatric epilepsy samples (Slick et al., 2006), the Plan/Organize scale was not related to HRQOL. Similarly, ratings on the Organization of Materials scale were not related to HRQOL. This suggests that difficulties with planning/organization, although common in severe epilepsy, do not translate into poor HRQOL unlike other executive functioning domains such as emotional control, inhibition, initiation, monitoring, shifting, and working memory.
We did not find that seizure frequency was a strong predictor of quality of life; this may relate to the fact that seizure frequency was recorded in terms of seizure counts rather than parent ratings of severity, as in some studies (Sabaz et al., 2001; Sabaz et al., 2003). Parent ratings may be useful in tapping overall severity of a seizure disorder, but they also introduce common method variance that increases the association between the two variables. Instead, we found that the neurological variables that were most predictive of HRQOL were signs of intractability such as medication load and number of failed AED trials. In the latter case, it is possible that increasing failure of medications may lead parents to adopt a “learned hopelessness” regarding their child's HRQOL, and suggests that repeated medication trial failures have a cumulative adverse impact on HRQOL which is not accounted for by seizure frequency alone. As well, we found that unlike the ICI, the HARCES was not associated with executive dysfunction. The HARCES provides information on the child's access to age-appropriate activities in daily life; therefore, the extent of activity limitations in children with epilepsy appears to be based more on neurological and epilepsy-related factors than on limitations imposed by executive functioning problems.
Some caveats deserve mention. These include the fact that our sample was a tertiary care center sample referred for neuropsychological assessment, which may have biased the sample toward greater severity of epilepsy and executive dysfunction. Although relevant for health professionals working in tertiary care centers, the findings may therefore not apply to community referrals, or to children with less severe forms of epilepsy. Secondly, our sample included a small group of children who had a prior history of epilepsy surgery, as well as children awaiting epilepsy surgery and a small group of children who were being treated with VNS. It is possible that HRQOL predictors might differ in these groups as a function of surgery status, or that predictors may differ in groups receiving different treatments for epilepsy (e.g., epilepsy surgery vs. VNS). Further studies using larger samples would be helpful to delineate the nature and extent of treatment-related changes in HRQOL predictors. In addition, it would be helpful to determine the degree to which executive dysfunction predicts HRQOL when additional factors are taken into account; ideally, these would include psychosocial, demographic, and family variables.
Based on the results of this study, we may now add executive dysfunction to the list of known risk factors for poor HRQOL in children with epilepsy, along with such factors as low adaptive behavior level, low IQ, intractability of the epilepsy syndrome, psychosocial difficulties, AEDs, low family income, and early age at epilepsy onset (Devinsky et al., 1999; Sabaz et al., 2001; Sherman et al., 2002; Buelow et al., 2003; Miller et al., 2003; Sabaz et al., 2003; Williams et al., 2003; Sillanpaa et al., 2004). Other questions remain unanswered, such as the relationship between executive dysfunction and behavioral disorders such as ADHD and affective disorders such as depression and anxiety in pediatric epilepsy. Studies on the relationship between executive dysfunction and self-reported HRQOL would also be of interest. The results also suggest new possibilities for screening and treatment research, including determining whether it would be useful to screen for executive dysfunction to identify children at risk for poor HRQOL to facilitate early intervention, and whether treatments aimed explicitly at improving executive dysfunction (e.g., stimulant medication, behavioral interventions) might improve HRQOL in children with epilepsy.