Statistical Analysis
An intent-to-treat analysis was conducted in patients
who had received at least one dose of study medication
and had a subsequent efficacy observation. Patients were
included in the safety analysis, if they had received one
dose of trial medication after randomization and had a
subsequent safety observation. The null hypothesis that
the expected mean weight change from baseline to the
end of 12 months of double-blind treatment did not differ
significantly between orlistat and sibutramine treatments
was tested using a three-way ANOVA and analysis
of covariance (ANCOVA) models [15]. Similar analyses
were applied to the other parameters. The statistical
significance of the independent effects of treatments
and weight loss on the other parameters was determined
by ANCOVA, using weight loss as the covariate. Statistical
analysis of data was performed by means of the SPSS
statistical software package for Window (version 11.0;Chicago); data are presented as meanSD. For all statistical
analysis, p