Evaluation of the multiple Assessment Subjective Shoulder Score (MASSS) : a pilot study

Introduction: This study compares an entirely patient reported modification of the Constant-Murley Score (CS) (the Multiple Assessment Subjective Shoulder Score [MASSS]), with the CS in symptomatic patients with non-instability shoulder pathology. Hypothesis: The hypothesis is that the sub scores and overall score of the MASSS will be well correlated and in high agreement to the corresponding scores of the CS. Materials and methods: The MASSS, the CS and adjusted-CS (a-CS) were administered to 74 patients. The MASSS replaced the strength and ROM domains of the CS with SANE scores. Correlations, differences, agreement, test-retest reliability, and internal consistency were assessed using Pearson's R test, paired t-test, Bland-Altman method, intraclass correlations (ICC) and Cronbach's alpha. Results: Correlation between MASSS and CS (0.834) and a-CS (0.824) were excellent. Total MASSS (54.0) was higher than CS (41.8) (p < 0.001) and a-CS (47.8) (p < 0.05). MASSS strength and ROM domains were higher than those of the CS (Strength: 13.4 v 5.3 [p < 0.001], ROM: 24.8 v 20.8 [p < 0.001]. There was a floor effect for the strength domain of the CS (46.7% minimum score), but not the MASSS (0% minimum score). MASSS internal consistency (0.626–0.734) was better than the CS (0.401–0.643). Test re-test reliability of the total MASSS (ICC 0.93) and the strength (ICC 0.90) and ROM (ICC 0.86) domains was excellent. Discussion: The MASSS has several advantages over the CS. Although the correlation of the total MASSS with the CS and a-CS was high, there was poor agreement in the strength and the ROM domains and therefore the MASSS value is not interchangeable with the CS, but rather should be considered as an alternative. Level of evidence: II; Prospective cohort study.