Reliability of Self-Reported Pre-Stroke Function

Abstract

P135 Purpose: The purpose of this study was to examine the reliability of self-reported pre-stroke function when reported within 28 days post-stroke and again at 90 days post-stroke. Methods: Sixty consecutive patients who participated in the Kansas City Stroke Registry were included in this analysis. Pre-stroke function was self-reported by participants using the Lawton IADL and SF-36 Physical Function Index at baseline (within 28 days of stroke onset) and again at 90 days post-stroke. The Intra-class Correlation Coefficient (ICC) was used to examine the reliability of self-reported pre-stroke physical function. The Orpington Prognostic Scale was used to characterize stroke severity. Results: Sixty subjects (24 males and 36 females) with an average age of 75±8.2 years were included in this study. Stroke severity at baseline was mild in 19, moderate in 37, and severe in 4. When compared to pre-stroke function reported at baseline using the Lawton IADL, prior function reported at 90 days was lower in 17 subjects, no change in 21 subjects, and higher in 22. Similarly, when compared to pre-stroke physical function reported at baseline using the SF-36 PFI, prior function reported at 90 days was lower in 18 subjects, no change in 11 subjects, and higher in 30. The calculated ICCs were .53 for both measures. Conclusions: Recall of pre-stroke function is not consistent across time.