OBJECTIVES:  While an increasing number of studies have examined patient preferences for diabetes medications, few studies have explored preference heterogeneity. We sought to understand patients’ preference heterogeneity in diabetes medications in this study and explore factors that potentially contributed to this variation. METHODS:  A nationally representative survey was conducted among patients with type II diabetes. The survey contained a discrete-choice experiment where each respondent was asked to choose between two hypothetical diabetes medications in 18 scenarios. These medications differed in six attributes: effectiveness in reducing hemoglobin A1c, duration of stable blood glucose levels, frequency of hypoglycemia, duration of nausea, treatment burden, and cost. Data were analyzed using a latent class conditional logit model to detect preference heterogeneity.  RESULTS: