Related Experiment Videos
Evaluation of proxy responses to the Stroke Impact Scale
Pamela W Duncan1, Sue Min Lai, Denise Tyler
1Brooks Center for Rehabilitation Studies, University of Florida, and Rehabilitation Outcomes Research Center, North Florida/South Georgia Department of Veteran Affairs, Gainesville, Fla 32610-0185, USA. pwduncan@hp.ufl.edu
Stroke
|November 2, 2002
Summary
Proxy reports on the Stroke Impact Scale (SIS) and SIS-16 are valid for stroke outcome assessment. Despite some differences in reporting severity, the biases are not clinically significant.
Area of Science:
- Neuroscience
- Rehabilitation Medicine
- Clinical Psychology
Background:
- Stroke patient outcomes are frequently assessed using patient self-reports.
- Proxy-patient agreement is crucial for understanding the reliability of these assessments when patients cannot self-report.
Purpose of the Study:
- To compare proxy and patient responses on the Stroke Impact Scale (SIS) and SIS-16.
- To estimate bias and evaluate the validity of proxy scores in stroke outcome assessment.
Main Methods:
- 287 patient-proxy pairs from the Kansas City Stroke Registry were assessed.
- Utilized various scales including SIS, SIS-16, modified Rankin Scale, and cognitive assessments.
- Proxy interviews were conducted within 7 days of patient interviews.
Main Results:
- Proxies rated stroke severity higher than patients on SIS and SIS-16, with bias increasing with stroke severity.
- Magnitude of bias was small (effect size -0.1 to 0.4), with good agreement (ICC 0.50-0.83).
- Concurrent validity with other functional measures was good to excellent for both patient and proxy scores.
Conclusions:
- Proxy-provided information is valid for assessing stroke outcomes.
- While significant differences exist between patient and proxy reports, biases are small and not clinically meaningful.
- Proxy reports can be reliably used in stroke research and clinical practice.