Our purpose was to determine if the glycemic index (GI) and proportion of carbohydrate absorbed as glucose (Pg) affected glycemic responses and occurrence of post-prandial hypoglycemia in subjects with type 1 diabetes treated with insulin lispro. Subjects (n=8) were studied on five separate occasions after 10-12 h overnight fasts following a standard dinner. After their morning insulin dose, subjects ate 50 g carbohydrate from a starchy food (Pg=1; mashed potato GI=83, white bread GI=71, spaghetti GI=41, barley GI=25) or pineapple juice (Pg=0.5; GI=46). Blood glucose was measured fasting and at 30 min intervals for 4 h. Glucose responses after different foods differed significantly from 30 to 180 min, with mean incremental area under the curve being closely related to GI (r=0.98, P<0.01). By multiple regression analysis, occurrence of post-prandial hypoglycemia was influenced (P<0.05) by subject and Pg. Time to hypoglycemia was affected by subject, fasting glucose, and GI. Thus, in subjects with type 1 diabetes treated with insulin lispro, GI predicts glycemic responses of carbohydrate foods. Pg may affect the occurrence of post-prandial hypoglycemia, while GI may affect its timing. Further studies using mixed meals are required to confirm how carbohydrate source affects glycemic responses and occurrence of hypoglycemia in normal meal setting.