Statistical Analysis of Series of N-of-1 Trials

The statistical analysis of series of n-of-1 trials in which the treatments are randomized in cycles within individuals is described. Version 9.4 (TS1M3) of the SAS software is used for the statistical analysis. The statistical methods demonstrated assume that the outcome variable is normally distributed. The examples presented are for a comparison between two treatments. Several methods appropriate for the estimation and inference on the overall treatment effect as well as on the individual treatment effects are compared and discussed.