ABSTRACT Introduction In recent years, the Ankylosing Spondylitis Performance Index (ASPI) has emerged as a new objective measure of physical function in patients with axial spondyloarthritis (axSpA). So far, only the relative interrater reliability of the ASPI test has been reported, but the absolute interrater reliability remains to be investigated. Objective This study investigated the interrater absolute and relative reliability of the ASPI test in patients with axSpA from a Nordic country. Methods Patients with axSpA were recruited from a Danish rheumatology outpatient clinic between October and November 2023. The participants performed the ASPI twice with two different raters, with a 15‐min pause between tests. Absolute and relative reliability were assessed. Results Fifty‐two participants were included. The standard error measurement for the overall ASPI test was 3.4%, and the minimal detectable difference was 9.5%. The participants performed the ASPI test faster with the second rater (mean 38.15 s (standard deviation, SD 12.16 s) compared to the first rater (40.92 s (SD 12.07 s)) with a statistically significant reduction in time of −2.77 s (SD 4.53 s) ( p < 0.001). The intraclass correlation coefficient of the ASPI test was 0.91 (95% CI 0.76–0.96). Conclusion The study showed high absolute interrater reliability with low interrater measurement error for single individuals and groups in patients with axSpA. The relative reliability was also high. However, ASPI demonstrated a learning effect at group level. This study offers clinically relevant guidance for interpreting differences in the ASPI scores when tests are repeated by different raters, relevant for real‐world clinical practice.
Rasmussen et al. (Thu,) studied this question.