Maintaining euglycemia (defined as glucose <110 mg/dL in intensive care unit settings and <180 mg/dL in non–intensive care unit settings by American College of Endocrinology Consensus Guidelines) has been shown to greatly reduce morbidity and mortality. This has created a paradigm shift in the inpatient setting where the goal is now to prevent hyperglycemia rather than treat hyperglycemia after it develops. Insulin, in both intravenous and subcutaneous regimens, is being used in the inpatient setting to achieve optimum glucose control. The clinical challenge with the use of insulin includes recognizing when to hold or decrease insulin and in which type of diabetic patient insulin should be adjusted in the nil per os (NPO) setting.

Considerable confusion has been caused by the previous diabetes classifications of insulin-dependent diabetes mellitus (IDDM) and non–insulin-dependent diabetes mellitus (NIDDM). Because of the obesity epidemic, NIDDM patients often progress to requiring insulin, causing health care providers to then refer to these patients as having IDDM. Yet, the NIDDM patient who progressed to requiring insulin is physiologically very different from the patient who was diagnosed initially as an IDDM patient. While both patients require insulin, the physiological cause of their insulin requirement and the response to exogenous insulin is significantly different.

In response to this issue, diabetes is now classified based on the etiology of the respective disease process. These classifications are as follows: