Community Hospital-based Stroke Programs: North Carolina, Oregon, and New York. III. Factors influencing survival after stroke: proportional hazards analysis of 4219 patients.

The possible effect of age, race, sex, consciousness upon admission, geographic location, and history of selected risk factors on the survival after stroke due to infarction or hemorrhage was determined using proportional hazards analysis (Cox regression). For each diagnostic category the most significant prognostic factor was consciousness upon admission. Increasing age, cardiac disease, or previous stroke also decreased the survival time of patients with infarctions. For patients with cerebral hemorrhage, no other variable was significant after control for consciousness level.