For each type of antibiotic, those used for ≥3 days were designated 'used' while all others were designated 'unused.' A dummy variable regression was subsequently performed.
Binary descriptive statistics and multiple variable regressions were done.
Candidate models for multiple-variable regression were those in which individual factors demonstrated an association that was statistically significant or approaching statistical significance, and the individual factors were not highly collinear.
Stepwise errors-in-variables regression was used to model the percentage of remaining tumour volume based on the pre-treatment volume, summary DCE-MRI statistics, and heterogeneity statistics using data from the first pre-treatment scan.
Frequencies, correlations and multi-variable linear regression were used to analyse the data.
Because the outcome variable was a count variable, poisson regression was performed [ 25].
All predictor variables in the regression were categorical.
Cases and controls were compared; significant variables from univariable regression were adjusted in a multivariable logistic regression model.
Single variable quantile regression was used to compare the median length of stay by visit type.
Single variable logistic regression was used to test for differences in the odds of admission/transfer/death between the injury and non-injury visits.
To the continuous variable, a regression was done to show the differences between the control and tests.
