To compare the pathophysiology of impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) in a more comprehensive and standardized fashion than has hitherto been done. We studied 21 individuals with isolated IFG (IFG/normal glucose tolerance [NGT]), 61 individuals with isolated IGT (normal fasting glucose [NFG]/IGT), and 240 healthy control subjects (NFG/NGT) by hyperglycemic clamps to determine first- and second-phase insulin release and insulin sensitivity. Homeostasis model assessment (HOMA) indexes of beta-cell function (HOMA-%B) and insulin resistance (HOMA-IR) were calculated from fasting plasma insulin and glucose concentrations. Compared with NFG/NGT, IFG/NGT had similar fasting insulin concentrations despite hyperglycemia; therefore, HOMA-IR was increased approximately 30% (P < 0.05), but clamp-determined insulin sensitivity was normal (P > 0.8). HOMA-%B and first-phase insulin responses were reduced approximately 35% (P < 0.002) and approximately 30% (P < 0.02), respectively, but second-phase insulin responses were normal (P > 0.5). NFG/IGT had normal HOMA-IR but approximately 15% decreased clamp-determined insulin sensitivity (P < 0.03). Furthermore, HOMA-%B was normal but both first-phase (P < 0.0003) and second-phase (P < 0.0001) insulin responses were reduced approximately 30%. IFG/NGT differed from NFG/IGT by having approximately 40% lower HOMA-%B (P < 0.012) and approximately 50% greater second-phase insulin responses (P < 0.005). Since first-phase insulin responses were similarly reduced in IFG/NGT and NFG/IGT, we conclude that IFG is due to impaired basal insulin secretion and preferential resistance of glucose production to suppression by insulin, as reflected by fasting hyperglycemia despite normal plasma insulin concentrations and increased HOMA-IR, whereas IGT mainly results from reduced second-phase insulin release and peripheral insulin resistance, as reflected by reduced clamp-determined insulin sensitivity.