Abstracts

Rationale: Several studies have demonstrated an association between behavioral/psychiatric disorders and childhood epilepsy. It is unclear, however, whether this association is dependent on or independent of the well-documented cognitive comorbidity that accompanies these epilepsies. Methods: Children with the Connecticut Study of Epilepsy (CSE, recruited 1993-1997) underwent comprehensive reassessments 8-9 years after the diagnosis of epilepsy (in 2002-2006). For those ≤16 years, this included cognitive testing with a Wechsler IQ test and behavioral assessment with the parent-reported Child Behavior Check List (CBCL). Similar-aged, neurotypical controls without epilepsy were obtained from among the siblings of cases and received the same assessment instruments. Cases with epilepsy divided into those with complicated (IQ<70, clear brain insult or abnormal exam) and uncomplicated (IQ≥70, all exams normal, i.e. neurotypical) groups. In this study, case group was comprised after excluding complicated cases and cases who were not seizure-free and on medication. Unmatched t-tests were used to test differences between controls and cases (seizure-free ≥ 5 years and off AEDs on Wechsler IQ test and CBCL scores. We also used the recommended clinical cut-points to create dichotomous variables reflecting clinically relevant CBCL problem scores. Multiple linear regression was used study the association of epilepsy and mean behavioral scores after adjustment for IQ. Multiple logistic regression was used to test the association of epilepsy with the number of subjects with/without clinically significant behavioral problems after adjustment for IQ. Results: The CBCL and Wechsler IQ test results of 301 cases and 156 controls were analyzed. The average age was 12.5±3.0 and 13.2±2.6 years in cases and controls (