Inter-rater and Intra-rater Reliability of the Cutting Movement Assessment Score Among Physical Therapists

Background Women’s basketball has a high incidence of anterior cruciate ligament (ACL) injuries, many of which occur during change of direction (COD) movements. Evaluating movement quality during these maneuvers can provide valuable insight into an athlete’s ACL injury risk. Purpose The primary aim of this study was to assess the reliability of the Cutting Movement Assessment Score (CMAS) in female basketball players when utilized by physical therapists (PTs) with varying levels of clinical experience and expertise. Study Design Cross-sectional study Methods Qualitative movement assessment was conducted using 2-dimensional video recording of a 45° change of direction (COD) task of 20 female basketball players and scored using the CMAS. Intra-rater and inter-rater reliability were assessed across two experts with ≥8 years of clinical experience and orthopaedic and sports clinical specializations, and one novice rater with <2 years of clinical experience and no board certifications. Intra- and inter-rater reliability for total CMAS scores were evaluated using intraclass correlation coefficients (ICCs). Agreement for individual CMAS items was assessed using percent agreement and kappa coefficients. Results The CMAS demonstrated excellent intra-rater reliability (ICC = 0.96) and moderate inter-rater reliability (ICC = 0.67) for total scores across all three raters. For individual CMAS items, intra-rater percentage agreement was excellent (80-100%), and kappa coefficients were moderate to excellent (k = 0.59-1.00). Inter-rater percentage agreement for individual items was moderate to excellent (60-90%), and kappa coefficients were slight to excellent (k = 0.12-0.79). Conclusion The CMAS can be a reliable tool for evaluating COD movement quality among PTs at various levels of clinical experience and expertise, although the identification of movement faults may have a relationship with years of clinical experience. The findings of this investigation suggest stronger intra-rater reliability than inter-rater reliability. Level of Evidence Level 3