[Intra- and interobserver reproducibility of the French version of the Constant-Murley shoulder assessment during rehabilitation after rotator cuff surgery]. / Etude de reproductibilité intra et inter-observateur de la version française du score de Constant au cours de la rééducation des coiffes opérées.

PURPOSE OF THE STUDY: The Constant-Murley scapular score is currently considered to be the gold standard for shoulder assessment in Europe. Few studies have examined the metrological qualities of this score. Our aim was to study the reliability and validity of the French version in a population undergoing reeducation after rotator cuff surgery. We wanted to determine how pertinent the score is during the reeducation phase (1-12 months after surgery). MATERIAL AND METHODS: Fifty-three patients volunteered to participate in this study. Shoulder assessment was performed by three observers. Intraobserver reproducibility was determined for 102 tests and two series of 32 and 56 tests were used to determine interobserver reproducibility. The internal coherence was studied on a sample of 61 tests. Three observers analyzed the apparent validity of the Constant Murley score. RESULTS: The correlations were satisfactory (intraobserver 0.96; interobserver 0.91 and 0.89 with the Spearman test) and sensitive (intraobserver 0.01; interobserver 0.07 and 0.01 with the Wilcoxon test). Despite satisfactory internal coherence (Cronbach alpha=0.75), the reproducibility of the overall score did not correspond necessarily to the reproducibility of the constituent scores. The measurement error might be related to patient- and observer-related interpretation variability. The apparent validity of the French version might be criticized for assessing rotator cuffs after surgery. CONCLUSION: A precise consensual protocol is needed for conducting the shoulder assessment and establishing the Constant-Murley score during the reeducation phase after rotator cuff surgery.