Reaching Performance Scale for Stroke - Test retest reliability, measurement error, concurrent and discriminant validity

S K Subramanian, M C Baniña, A Turolla, M F Levin

Research output: Contribution to journalArticlepeer-review


INTRODUCTION: Post-stroke upper limb motor improvement can be better quantified by describing movement patterns characterizing movement quality and use of compensations. Movement patterns can be described using both kinematic and clinical outcomes. One clinical outcome that assesses movement quality and compensations used for reaching a Close (18pts) and Far target (18pts) is the Reaching Performance Scale for Stroke (RPSS).

OBJECTIVE: We estimated the pilot test-retest reliability and validity (concurrent, discriminant) of the RPSS in individuals with chronic stroke.

DESIGN: Retrospective data analysis.

SETTING: Research laboratory.

PARTICIPANTS: Seventy-two individuals with upper limb hemiparesis ≥6 months prior to participation.

INTERVENTION: Not Applicable.

MAIN OUTCOME MEASURE: RPSS Close and Far Target scores. Intraclass correlation coefficients (ICCs) helped assess pilot test-retest reliability on a subset of 14 participants. Concurrent validity was assessed for individual RPSS items with corresponding kinematic outcomes (trunk displacement, shoulder flexion, shoulder horizontal adduction, elbow extension, trajectory straightness) using Pearson's correlations. We also ran multiple regression analyses with the RPSS total scores and used kinematic outcomes as the criterion standard. Logistic regression analyses estimated discriminant validity. We divided participants into two groups based on the FMA scores (mild:≥50/66; moderate-to-severe:≤49/66).

RESULTS: Test-retest reliability was excellent for close (ICC = 0.98, 95%CI:0.94-0.99) and far targets (ICC = 0.98, 95%CI:0.95-0.99). Individual RPSS items for both targets were mildly to moderately correlated with corresponding kinematic values. A combination of trajectory straightness, elbow extension and trunk displacement explained majority of the variance in RPSS scores (47%) for both targets. The RPSS scores discriminated between individuals with mild and moderate-to-severe motor impairment for both Close (ExpB = 3.33, P < 0.001; 95%CI:1.70-6.52) and Far targets (ExpB = 2.59, P < 0.001, 95%CI:1.65-4.07). Cutoff points for transition between groups were 15.5 (close target) and 14 (far target).

CONCLUSION: The RPSS is a valid clinical measure with excellent pilot results of test-retest reliability for assessing movement patterns and compensations used for reaching.

Original languageEnglish
JournalPM and R
Publication statusE-pub ahead of print - Mar 6 2021


Dive into the research topics of 'Reaching Performance Scale for Stroke - Test retest reliability, measurement error, concurrent and discriminant validity'. Together they form a unique fingerprint.

Cite this