Twenty-six children with intellectual disability and six normal children, all suffering from refractory partial seizures, received open-label gabapentin (range = 10-50 mg kg(-1) day(-1); mean = 26.7 mg kg(-1) day(-1) as an add-on medication to their antiepileptic drug regimen. Mean seizure frequency during baseline was 9.5 seizures per week. Both groups had a significant reduction in seizure frequency. Response scores and response ratios did not differ between the intellectually disabled and normal groups (1.67+/-0.67 and 1.25+/-0.69, P = 0.697, and -0.400+/-0.089 and -0.283+/-0.159, P = 0.961, respectively). Behavioural side-effects were more likely to occur in patients with intellectual disability in comparison with the mentally normal group (P = 0.0107). In the present patient population, patients younger than 10 years of age, all of whom had intellectual disability, were more likely to have side-effects than those older than 10 years of age. Observed adverse effects, which were generally mild, occurred in patients with baseline intellectual disability, attention deficit disorder and behavioural problems. Behavioural adverse effects warranted discontinuation of the medication in only three patients. The severity of intellectual disability (mild versus moderate or severe) did not affect the extent of the response or the occurrence of side-effects. It is concluded that gabapentin is equally effective as an add-on medication against partial seizures in patients with or without intellectual disability. However, children with intellectual disability who also are less than 10 years of age with baseline attention deficit appear to be at a higher risk of behavioural side-effects.