Intra- and Interrater Reliability of Remote Assessment of Transfers by Wheelchair Users Using the Transfer Assessment Instrument (Version 4.0).

Abstract

To evaluate the reliability, standard error of measurement, minimum detectable change, and item-level consistency of the Transfer Assessment Instrument (TAI) version 4.0 when used to evaluate transfer quality remotely. Participants transferred from their wheelchair to a mat table (transfer 1), repeated this after a 10-minute delay to assess intrarater reliability (transfer 2), and repeated this 1-2 days later to assess test-retest reliability (transfer 3). Each transfer was scored in person by 4 raters and asynchronously by a remote clinician rater. 2017 National Veterans Wheelchair Games. Convenience sample of 44 full-time wheelchair users (N=44). Not applicable. TAI total score, subscores (wheelchair setup, body setup, flight/landing), and item scores (15 items). Moderate to excellent reliability was found when scoring remotely for TAI total and subscores for intrarater (intraclass correlation coefficient (ICC(3,1)=0.687-0.854), test-retest (ICC(3,1)=0.695-0.836), and interrater reliability (ICC(3,5)=0.746-0.962). Remote rater total score and flight/landing subscore were greater (indicating higher transfer quality) compared to the average in-person raters (P=.021 and P=.005, respectively). There were no differences between transfers 1-3 in remote rater scores. Item-level percentage agreement between the remote rater and in-person exceeded the 75% cutoff for clinical utility for all items. The TAI is a reliable outcome measure for assessing transfer technique remotely.