RESUMEN
We assessed a screening instrument, adapted from a model suggested by WHO, aimed to perform population studies on the prevalence of cerebrovascular disease in Chile. Sixty-two subjects, 31 with cerebrovascular diseases and 31 without, were asked about symptoms and requested to do simple movements by trained interviewers. The results of the instrument were compared with a neurological examination performed by two specialists. Global sensitivity and specificity of the instrument, using WHO evaluation criteria, were 100 and 38.7 percent respectively. When three or more symptoms and one positive sign were considered as cutoff points, global specificity increased to 61 percent and sensitivity decreased to 93 percent. It is concluded that the present instrument is highly sensitive but lacks specificity