Two important challenges are inherent in the design of studies using prescription data from electronic health records: how to define the minimum level of adherence that would qualify as "continuous drug use" and how to handle stockpiling of medications. Generally, the sensitivity of a study's conclusions to these design choices is not analyzed. In our study, covariate adjusted Cox models were used to compare persistence and durability with respect to three common oral antidiabetic therapies in a cohort of 12,697 incident users. Assuming 50% stockpiling, sulfonylurea therapy, as compared with metformin, showed a significantly lower risk of nonpersistence (changing or stopping therapy) when no gap days were allowed (HR 0.95, P = 0.032), no significant difference when 14 gap days were allowed (HR 0.99, P = 0.536), and significantly greater risk of nonpersistence when 30 gap days were allowed (HR 1.05, P = 0.046). All the drug comparisons showed statistically significant effects in both directions, the risk of nonpersistence increasing or decreasing depending on the design parameters.