Each outcome is analyzed using appropriate statistical methods. For example, the Kaplan-Meier estimates of the time to discontinuation for intervention and control groups are used to graphically depict patient persistency. The between-group statistical comparison in this case is performed with the log-rank test (p-value: p<=0.001).



Another way to visualize and analyze program results is to build and compare cumulative distribution curves of the mean number of pills per patient consumed over the study period. The figure below shows an example of such analysis conducted at one year after program inception.