22.3. Data analysis
To compare the traction group and the control group,
Statistical analysis was based on the intention-to-treatprinciple, and p-values less than 0.05 were considered
significant. We used multiple imputations to handle
missing data. To impute the missing data we constructed
multiple regression models including variables
potentially related to the fact that the data were
missing and also variables correlated with that outcome.
We used Stata (Stata Corp, College Station,
Texas, USA).Analysis of covariance (ANCOVA) at
two follow-up points (after 10 weeks of treatment and
at follow-up of six months) was performed for all variables,
the baseline value of the outcome as covariates
was used to assess between group differences.(baseline
outcome in the mode = baseline value – overall mean
baseline value).