Analysis of the effects of treatment. The statistical analysis of our study will follow intention-to treat principles.36 The normality of the data will be tested by visual inspection of histograms, and the characterization of the participants will be calculated using descriptive
statistical tests. The between-group differences (effects of treatment) and their respective 95% confidence intervals will be calculated by constructing mixed linear models 37 using interaction terms of treatment groups versus time. We will conduct a secondary exploratory analysis to assess whether patients classified as having derangement syndrome have a better response to the McKenzie method (compared with placebo) than those with other classifications. For this assessment, we will use a 3-way interaction for group, time, and classification. For all of these analyses, we will use the IBM SPSS software package, version 19 (IBM Corp, Armonk, New York).