Dennis JM, et al. - For patients initiating type 2 diabetes therapy, researchers applied joint longitudinal–survival models to assess relationships between drug response (longitudinal outcome) and the risk of side effects (survival outcome). For this investigation, 4,351 participants were randomized to metformin (MFN), sulfonylurea (SU), or thiazolidinedione (TZD) therapy in the A Diabetes Outcome Progression Trial (ADOPT) drug efficacy trial. A greater response was correlated with an increased risk of edema but not fracture with TZD. Greater current response was related to an increased risk of hypoglycemia with SU. Joint modeling provides a useful framework for assessing the relationship between drug response and the risk of developing side effects. The study suggests the potential for joint modeling to assess the benefits and risks of drugs in drug development and precision medicine research.