Abstract
Data on 515 consecutive patients registered with the Oxfordshire Community Stroke Project were used to compare the characteristics of those patients who were admitted to hospital within one month after their first stroke with those who remained in the community during that time. Twenty eight patients had their stroke while in hospital for other conditions, and of the remaining 487, 266 were admitted. Though patients with a severe neurological deficit were significantly more likely to be admitted, 47 out of 202 such patients were managed in the community. In a substudy of 162 consecutive patients the general practitioners' reasons for either arranging admission to hospital or continuing with community care in the first week after the stroke were ascertained. Sixty patients were admitted. The only reason for admission was diagnostic uncertainty in five cases (though this was a contributing factor in 25) and to provide nursing or general, non-medical care in 25. Patients who lived alone were more likely to be admitted. All 12 patients who presented directly to the casualty department were admitted, though only five had had a severe stroke. A stroke service that provides a facility for rapid outpatient and domiciliary diagnosis as well as a rapidly acting domiciliary nursing team might reduce the number of patients with stroke admitted to hospital without adversely affecting the quality of patient care: this should be properly evaluated.