In a recent publication in the Archives, Shurraw et al1 analyzed 23 296 diabetic patients with chronic kidney disease (CKD) and demonstrated a U-shaped relationship between glycemic control and excess mortality. Their use of extensive and sophisticated statistical analyses is highly praiseworthy. The acknowledged limitations include its retrospective observational nature, the use of first hemoglobin A1c (HbA1c) measurements for patient classification despite use of various sensitivity analyses, and the nonevaluation of the impact of blood pressure control on patient outcomes.1 While agreeing with the general conclusions, the following concerns must be recognized.