At least 90% of the 12 to 15 million persons with diabetes mellitus in the United States, half of whose condition remains undiagnosed, have type 2 diabetes. Type 2 diabetes is preceded by a long period of impaired glucose tolerance, a reversible metabolic state associated with increased prevalence of macrovascular complications. Thus, at the time of diagnosis, long-term complications have developed in almost one fourth of patients. Susceptibility to type 2 diabetes requires genetic (most likely polygenic) and acquired factors, and its pathogenesis involves an interplay of progressive insulin resistance and beta-cell failure. The ideal treatment of type 2 diabetes should reverse insulin resistance and beta-cell dysfunction in most treated patients and prevent, delay, or reverse long-term complications. Current strategies are aimed at amelioration of insulin resistance (diet, exercise, weight loss, and metformin and troglitazone therapy), augmentation of insulin supply (sulfonylurea and insulin therapy), or limitation of postprandial hyperglycemia (acarbose therapy). Future therapies probably will target (1) insulin resistance, using a multifaceted approach; (2) hepatic glucose production, using gluconeogenesis inhibitors; (3) excess nonesterified fatty acid production, using lipolysis inhibitors; and (4) fat oxidation, using carnitine palmitoyltransferase I and II inhibitors. Attempts also could be made to stimulate energy expenditure and increase nonoxidative glucose disposal by means of beta 3-adrenoceptor agonists. One promising strategy is an attack on multiple pathophysiological processes by combining antidiabetic agents with disparate mechanisms of action. Thus, we now have unprecedented resources for drug therapy for diabetes, with great opportunity for innovative combinations. It is hoped that these expanded choices will provide the tools necessary for a more efficient management of type 2 diabetes and prevention of its long-term complications.