The article reviews the effect of diabetes on the pharmacokinetics and pharmacodynamics of drugs in humans. For most drugs which cross the gastrointestinal wall by passive diffusion, oral absorption is unlikely to be affected by diabetes, although a delay in the absorption of tolazamide and a decrease in the extent of absorption of ampicillin have been reported. Subcutaneous absorption of insulin is more rapid in diabetic patients, whereas the intramuscular absorption of several drugs is slower. The binding of a number of drugs in the blood is reduced in diabetes, which may be due to glycosylation of plasma proteins or displacement by plasma free fatty acids, the level of which is increased in diabetic patients. Plasma concentrations of albumin and alpha 1-acid glycoprotein do not appear to be changed by the disease. The distribution of drugs with little or no binding in the blood is generally not altered, although the volume of distribution of phenazone (antipyrine) is reduced by 20% in insulin-dependent diabetes mellitus (IDDM). In contrast to animal studies, the metabolic clearance of most drugs in humans appears to be unaffected or slightly reduced by the disease. The presence of fatty liver in non-insulin-dependent diabetes mellitus (NIDDM) may contribute to a reduced hepatic clearance, whereas decreased binding in the blood may cause an increase in clearance. The effect of diabetes on hepatic blood flow in humans appears to be unknown. Diabetes affects kidney function in a significant number of diabetic patients. During the first 10 years after the onset of the disease, glomerular filtration is elevated in these patients. Thus, the renal clearance of a number of antibiotics has been shown to be increased in diabetic children. As the disease progresses, renal function is impaired and glomerular function declines from the initial elevated state. In diabetic adults the renal clearance of drugs either is comparable with that found in nondiabetic individuals or is reduced. A limited number of studies have been conducted comparing the dose-response of cardiovascular drugs in diabetic patients with that in nondiabetic controls. Decreased, increased and unchanged responses have been reported. It is apparent that in some cases an altered response may be observed for a drug when administered to a diabetic patient compared with a similar nondiabetic individual. At the present time, it is not possible to ascertain whether these studies reflect true pharmacodynamic changes or merely alterations in pharmacokinetics.(ABSTRACT TRUNCATED AT 400 WORDS)