Behavioral disorders in children with epilepsy: Relation to clinical and demographic variables

This study examines the prevalence of behavioral disorders in a consecutive series of 92 pediatric epilepsy subjects at Miami Children's Hospital, to determine the relationship of these disorders to clinical and psychosocial characteristics of the patients. Patients were divided into two groups: subjects with intractable epilepsy (n=57) and subjects with non-intractable epilepsy (n=35). All medical-neurological information was generated by the treating pediatric epileptologist and included clinical information such as seizure type, seizure frequency, seizure focus, and EEG and MRI findings. The presence of behavioral disorders was examined using the Child Behavioral Checklist (CBCL). Descriptive and frequency based statistics were obtained for the demographic and clinical variables for the entire sample, and for the intractable and non-intractable subgroups independently. A linear regression procedure was performed for the entire sample to determine which clinical/demographic variables were significantly related to the CBCL subscale scores The same procedure was also conducted for the two patient subgroups independently. Differences in demographic, clinical, and behavioral variables between the two patient subgroups were analyzed using t-test or non-parametric comparisons. Results revealed a high frequency of behavioral pathology for the entire sample ranging from 9-10% (acting out and aggressive behaviors) to 30-36% (problems with attention, thinking, and social-interpersonal problems). Evaluation of the clinical variables revealed that for 9 of the 11 subtests intractability was the only clinical variable that accounted for a significant portion of the variance in test scores (p<0001). Clinical variables such as seizure type and interictal EEG abnormalities were significantly related to social, attentional and behavioral problems for the intractable group. For the Non-Intractable group, the relationship between clinical variables and the specific subscales was not significant. Comparison of the two groups revealed that, for 9 of the 11 CBCL subscales, the intractable group demonstrated significantly greater behavioral pathology. This study revealed that, as a group, children with epilepsy have a higher incidence of behavioral disorders than what has been documented for the general population. Additionally, these findings indicate that the presence of intractable epilepsy represents a significant risk factor for behavioral pathology and that, for this group, there are select clinical variables that are related to these disorders of behavior.