Abstracts

Rationale: Children with seizure disorders are at risk for slower social development. Existing data indicate that weaknesses in social skills during childhood are fairly mild and may be largely related to intellectual and learning problems. Past studies, however, have primarily relied on parent report measures. This longitudinal study compared social skills in a group of intellectually normal children with epilepsy to those of a group of neurologically normal children, utilizing both parent and teacher report measures. Methods: Forty-two children who had been diagnosed with epilepsy in the previous six months were followed over a year and compared to a group of 22 normal controls. The two groups were not significantly different in age (M=8.68 years), gender or SES. Seizure types included generalized absence (22%), generalized tonic-clonic (27%), complex-partial without generalization (24%) and complex-partial with generalization (27%). Although the average IQ of both groups was in the average range, the IQ of the control group was significantly higher so IQ was used as a covariate in each repeated measure analysis of variance. The Social Skills Scale from the Personality Inventory for Children (PIC) and the Socialization Domain Scale from the Vineland were employed as parent measures of social skills. The Social Skills Scale from the Comprehensive Behavior Rating Scale for Children (CBRSC)was used as the teacher report measure. Results: At each time point correlations between the parent and teacher report measures were significant but moderate in size (r=.29 to .50). Repeated measures ANOVA did not reveal either a group x time interaction or a main effect of time for any of the social skills measures. Analyses revealed that children with epilepsy at both time points had more poorly developed social skills as measured by both teacher report (p<.05)and maternal report from the Vineland (p<.01). Although a similar pattern was evident from maternal responses to the PIC, the differences were not statistically significant. Analysis of variance did not reveal a significant association between seizure type and social skills assessed by any measure. Seizure frequency between the two times points was correlated with maternal report of poor social skills on the Socialization Scale of the Vineland but was not associated with scores from the PIC or the CBRSC. Range of seizure frequency in this sample, however, was too limited to provide a good test of the impact of seizure frequency on social competence. Conclusions: This cohort of children with epilepsy exhibited problems with social development that were evident by parent and teacher report. Over time, there was agreement between parent and teacher judgements of social competence. Differences in social adjustment were noted despite controlling for general level of intellectual functioning. Problems with social adjustment were noted within 6 months of diagnosis and persisted. Seizure type was unrelated to maternal perceptions of social skills.