Evaluating the psychometric properties of the Short Physical Performance Battery in community-dwelling individuals with stroke

Purpose: This study evaluated the psychometric properties of the Short Physical Performance Battery (SPPB) in individuals with stroke using Rasch analysis. We assessed its structural validity, measurement accuracy, invariance, test–retest reliability, and convergent validity with the Fugl-Meyer Assessment for the Lower Extremity (FMA-LE), Berg Balance Scale (BBS), and Timed Up and Go Test (TUG). Materials and methods: We recruited 152 stroke participants. Rasch analysis was performed using Winsteps. Convergent validity was assessed using Spearman’s rho (ρ), discriminative power was assessed via receiver operating characteristic curves. Test–retest reliability was assessed using intraclass correlation coefficient (ICC_3,1). Results: The median SPPB score was 7.0. The scale demonstrated unidimensionality, with 69.9% of variance attributed to the primary factor. Infit and outfit statistics ranged from 0.6 to 1.4. The gap between thresholds for sit-to-stand task categories 1 and 2 was substantial. Correlations ranged from moderate (ρ = 0.57, FMA-LE) to very strong (ρ = −0.86, TUG; ρ = 0.85, BBS). An SPPB score ≤6 indicated high fall risk with limited validity but showed very good to excellent test–retest reliability (ICCs > 0.80). Conclusions: SPPB is a valid, reliable tool for assessing physical function in individuals with stroke. However, clinicians should note potential limitations in the scale’s ability to fully capture true ability levels.Implications for rehabilitation The Short Physical Performance Battery (SPPB) is a reliable and valid tool for evaluating lower-limb physical function in people with stroke. Due to its limited ability to predict falls, clinicians should pair the SPPB with other assessments for a more accurate evaluation of fall risk. SPPB scores are converted to interval-level measures for researchers’ and clinicians’ reference.