with epilepsy’ SIR – We read with interest the article by Hoie et al.1 and note that the highest level of problems were found in the symptomatic and probable symptomatic group. We would like to suggest that one contributory cause to these psychosocial problems may be problems of social understanding, i.e. lack of understanding of the thoughts and intentions of others in interpersonal interactions. Problems of communication and social understanding in children with refractory focal epilepsy have been well documented.2,3 However, samples including children with well-controlled seizures and a probable symptomatic (cryptogenic) diagnosis have received less research attention. We invited 104 parents of children between the ages of 4 and 16 years within the caseload of the epilepsy nurse at the Royal Preston Hospital, Preston, UK, to complete the Children’s Communication Checklist, 2nd edition (CCC-II).4 These cases were selected such that the child was in mainstream education and the diagnosis was either idiopathic or probable symptomatic epilepsy. The CCC-II is a validated and standardized instrument for screening for clinically relevant communication problems, including specific language impairment and autistic spectrum disorder. Thirty-three questionnaires were returned, although only 31 could be used due to missing responses. The cases were categorized as focal (n=16; seven females, nine males), generalized (n=12; seven females, five males), or undetermined (n=3; two females, one male) based on information in the medical file of each child. Clinical presentation and electroencephalogram recordings formed the basis of diagnosis, with magentic resonance imaging data available for most of the sample. There was no significant difference between focal and generalized groups in terms of chronological age, but there was a trend for the focal group to have an earlier age at onset than the generalized group (focal: mean age 4y 6m [SD 2y 6m]; generalized: mean age 6y 9m [SD 3y]; t(26)=1.9, p=0.07), so age at onset was entered as a covariate when comparing the two groups on questionnaire data. Seventy-five per cent of the focal group was on monotherapy, with 69% of the group achieving seizure control. Fifty-eight per cent of the generalized group was on monotherapy and 75% of the group had achieved seizure control. The focal group scored significantly lower than non-clinical norms on all 10 subscales of the CCC-II (ranging from measures of syntax to social interaction and interests). They also scored significantly lower than non-clinical norms (scaled normed score=82) on a summary measure, the General Communication Composite (GCC; mean 48.5 [SD 29.0]; t=4.6, p<0.001). None of the differences between the generalized epilepsy group and non-clinical norms was significant, although the generalized group was heterogeneous, containing some children that scored very poorly, and others who showed superior performance. There were significant differences between the generalized and focal group for speech (F[1,25]=6.1, η2=0.20, p<0.05), syntax (F[1,25]= 5.5, η2=0.18, p<0.05), and stereotyped use of language (F[1, 25]=4.5, η2=0.15, p<0.05). GCC scores were also significantly higher for the generalized group relative to the focal group (F[1, 25]=4.3, η2=0.15, p<0.05). There was no other significant difference (possibly due to limited power), and age of onset was not significant as a factor in any comparison. Although there was clearly self-selection in the sample, this cannot fully account for the poor performance of the focal group, as the generalized group did not show this poor profile. Despite the preliminary nature of these data, it does suggest that clinicians and educators need to be aware that children with focal epilepsies may be particularly vulnerable to problems of social understanding and communication, irrespective of whether there is seizure control.