Statistical analyses were performed in Stata (Version 16.1; StataCorp LP). Descriptive statistics including mean and standard deviation were calculated. Intraclass correlation coefficients (ICCs) were calculated between reviewers for both total SOAS score and each individual criterion. Individual SOAS subscores that were not found to have a significant ICC were not included in further analyses. The Spearman rank correlation coefficient (r) was used to assess relationships between pairs of continuous variables, including total SOAS scores and patient variables, total SOAS scores and PROMIS-UE scores, and individual SOAS criterion scores and PROMIS-UE scores. The point biserial correlation was used to compare sex with PROMIS-UE and SOAS scores. A multivariate linear regression was performed with the independent variables of patient age, sex, BMI, and total SOAS score or subscore and the dependent variable of PROMIS-UE score, with P < .0125 considered significant after application of the Bonferroni correction with m = 4 comparisons. Statistical significance was defined as P < .05 for all other tests.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.