Purpose <p>To perform a systematic review with meta-analysis on reliability of Short Physical Performance Battery (SPPB) in adult patients.</p> Methods <p>Five databases were queried until March 2025 to identify studies evaluating SPPB intra- and inter-rater reliability and measurement error. Two independent reviewers performed study selection, data extrapolation and assessment of methodological quality and quality of evidence (QoE). Random effect meta-analyses were performed when at least two studies assessed the same property of the same subscale.</p> Results <p>Twenty-two studies were included. Regarding total score, intra-rater (ICC = 0.88) and inter-rater reliability (ICC = 0.86) were rated sufficiently with high QoE, while the measurement error (MDC = 1.90) was rated insufficiently with high QoE. Regarding the balance subscale, intra-rater (ICC = 0.69) and inter-rater reliability (ICC = 0.77) were rated inconsistently with low QoE, while the measurement error (MDC = 1.32) was rated as indeterminate with high QoE. Regarding the walking subscale, intra-rater (ICC = 0.88) and inter-rater reliability (ICC = 0.86) were rated sufficiently with moderate and high QoE, respectively, while the measurement error (MDC = 0.51 points) was rated indeterminately with moderate QoE. Regarding the chair subscale, intra-rater (ICC = 0.83) and inter-rater reliability (ICC = 0.94) were rated sufficiently with moderate and high QoE, respectively. In contrast, the measurement error (MDC = 2.14 points, equal to 53.4% of the total score) was rated indeterminately with high QoE.</p> Conclusion <p>Intra- and inter-reliability of the SPPB were supported for assessing physical performance in different populations. Measurement error of each subscale (except for the walking subscale) is higher than the measurement error of the total score; these results suggest using the total score in clinical practice.</p> <p><i>PROSPERO registration:</i> CRD420251003320.</p>