Cerebral malaria is one of the severe manifestations of falciparum malaria. It has a high fatality rate and survivors are at risk of a variety of neurological sequelae. The characteristics and determinants of neurological sequelae have been extensively studied in children aged under five years. Survivors who are over five years are known to develop sequelae and indications are that they may be at higher risk. This study set out to identify the sequelae that are present in children and adults in Uganda and to examine if there were differences in the occurrence, natural history and determinants of deficits. The study employed a multi-centre design using study hospitals that were located in malaria epidemic-prone areas in Uganda so as to enrol patients with a wider age distribution than would have been the case in malaria endemic sites. One hundred patients were enrolled between 1 January 2002 and 31st July 2003. The proportion of cerebral malaria among malaria admissions was 1.9% (72/3789) in children and 1.8% (28/1592) in adults admitted to the study hospitals. The case fatality ratio was low, 10%, and young children under five years of age at the time of admission had the highest mortality. Thirty-five survivors were detected to have neurological sequelae of which 63% were of the motor-sensory kind (22 survivors), 20% of the cognitive kind (7 survivors) and 17% had both kinds (6 survivors). The proportion of survivors with neurological sequelae was slightly higher in adults, 42% (11/26) than children, 38% (24/64). The more striking age difference was the frequency of sequelae among survivors aged under five, 28% (11/40), and those aged 5 to 9 years, 54% (13/24). Deficits were present in twenty-three survivors at discharge (motor-sensory sequelae only) and nineteen survivors that attended review visits (motor-sensory and cognitive sequelae). Deficits involving muscle tone (detected in 18 survivors), tendon reflexes (14 survivors) and muscle power (12 survivors) were the commonest at discharge, occurring in combination. Impaired muscle strength (5 cases) was the commonest motor-sensory sequela detected during follow-up. The commonest cognitive deficit either reported by mothers of survivors or detected by the cognitive tool was memory impairment which was found in 10 survivors. The natural history of sequelae did not differ between children and adults. Mortality was independently associated with quinine use within 48 hours prior to admission, seizures while on the ward and parasite count on admission. Quinine use prior to admission, duration of coma, duration with a history of fever plus days in hospital with a high temperature and splenomegaly were independently associated with early-onset deficits. The detection and development of late-onset deficits was associated with duration of elevated body temperature, duration taken to control seizures and the presence of neurological deficits at discharge. It was not possible to examine the determinants for the outcomes in children and adults separately.