Design of a global inter-rater reliability study of the poststroke spasticity referral tool.

Abstract

Background: Poststroke spasticity (PSS) occurs in =58% of stroke survivors. The simple, 1-page PSS Referral Tool was developed to facilitate PSS early identification and referral by clinicians involved in stroke rehabilitation. Red, yellow, and green indicate urgent referral to a spasticity specialist, routine referral with a recommendation for multidisciplinary team consult, or periodic monitoring, respectively. Objective(s): To validate the PSS Referral Tool use in clinical practice, an inter-rater reliability (IRR) study will be conducted. Method(s): This 3-part observational, prospective study will use the PSS Referral Tool to assess and classify patients up to 1 year poststroke with varying degrees and types of spasticity. Part A will consist of investigators at 2 sites recording 30 clinical assessment videos using a semi-structured script based on PSS Referral Tool categories. In Part B, 4 clinicians will select 15 videos (5 each for red, yellow, and green) for IRR assessment. In Part C, stroke rehabilitation clinicians recruited from 6 global regions will classify 5 randomly assigned videos into appropriate categories in 3 separate sessions. Result(s): This study will report the intraclass correlation coefficient (ICC) for IRR of the PSS Referral Tool, using a 2-way random effect model ICC with 95% confidence intervals. Sample size calculation for 80% power (alpha = .05) determined a need to recruit 270 clinicians stratified by 2 categories or 540 for 4 categories of PSS Referral Tool experience. ICC values range from 0.0 to 1.0, with higher scores indicating a more stable instrument. Conclusion(s): This study aims to ensure that the PSS Referral Tool possesses sufficient sensitivity and specificity for meaningful use by clinicians regardless of experience.