Medication non-adherence is a widespread problem that is associated with worse health outcomes and increased healthcare costs. Although many measures of adherence have been developed, their usage is not consistent across studies and statistical methods for analyzing these measures have not been rigorously evaluated. Using Proportion of Days Covered (PDC), a commonly used adherence measure, we illustrated the importance of defining the PDC calculation method and study cohort by sensitivity analyses. We also performed simulation studies to investigate properties of three statistical models. Our sensitivity analysis showed that parameter estimates can vary greatly depending on the rules for determining the study end date in calculating PDC, or the minimum number of fills in defining the cohort. In simulation studies, logistic regression had lower power than ordinal logistic and negative binomial regression models. Ordinal logistic regression is robust to distributional assumptions with greater power, but covariates affecting the observation window should be adjusted to reserve type I error rate. We also provided a recommendation for defining the observation window in calculating PDC.