Statistical analyses were performed. Chi-square tests were used to assess univariate association between independent variables and reported musculoskeletal symptoms. Multiple logistic regression analysis was performed for each of outcome retaining the variables (individual and perceived physical and psychological work demands) in the models to adjust for potential confounding. In the regression analysis, if the P value of Chi-square test for assessing association between the variables and reported symptoms was ≤0.25, the variable was included in the regression model of that region (inclusion criteria).[8] For each body region, this procedure was performed for all individual and perceived physical and psychological work demands variables.