Persisting consequences of stroke measured by the Stroke Impact Scale

The purpose of this study was to compare disability and quality of life as measured by the Stroke Impact Scale (SIS) of stroke patients deemed recovered (Barthel Index > or =95) with 2 stroke-free populations of community-dwelling elderly.

Eighty-one stroke patients who participated in the Kansas City Stroke Registry and achieved a Barthel Index of > or =95 at 3 months after stroke and 246 stroke-free subjects enrolled in the Community Elders Study were enrolled in this study. The Community Elders Study group was further divided into 2 groups, those recruited from the Department of Veterans Affairs Health System (VA) and a those from a local health maintenance organization (HMO). Stroke patients were administered the SIS approximately 90 days after stroke, and the stroke-free community dwellers were administered a version of the SIS adapted for nonstroke subjects, the Health Impact Scale (HIS). A general linear model was used to examine differences in health outcomes measured by the SIS or HIS between the KCSR stroke patients and VA and HMO community-dwelling elders after controlling for medical comorbidities and demographics.

Kansas City Stroke Registry participants were significantly older than the community study groups (P=0.0052). Selected medical conditions were similar among the 3 study groups. Old age and a history of diabetes mellitus were more likely to be associated with more deficits and poor quality of life. In stroke patients deemed recovered, stroke still affected hand function, activities and independent activities of daily living, participation, and overall physical function compared with the stroke-free community dwellers in the HMO health system even after adjustment for age and diabetes status. Stroke-free community dwellers in the VA health system also had worse social participation than the stroke-free community dwellers in the HMO health system.

Research and clinicians have consistently underestimated the impact of stroke with the Barthel Index. This has major implications for the design of therapeutic trial designs and adequate assessments of social and economic sequelae of stroke.