Computerized management of diabetes is the use of information technology to improve diabetic patient outcomes. The computer can be used to provide educational information to patients and facilitate the storage and transmittal of clinical data between patients and clinicians. The objective of this paper was to evaluate computerized management of diabetes in changing the health outcomes. Clinical trial reports were identified through systematic electronic database and manual searches. Four eligibility criteria were applied: diabetes clinical area; prospective, contemporaneously controlled clinical trial with random assignment of the intervention; computer generated information for patients in the intervention group and no similar intervention in the control group; and measurement of effect on the outcome of care (health status, social functioning, patient/family satisfaction). Data were abstracted using a standardized abstraction form and the quality of methodology was scored. Of 15 eligible clinical trials, 12 (80%) reported positive outcomes or significant benefits. A total of 48 outcome measures were reported, an average of 3.2/study. Significantly improved clinical outcomes included Hemoglobin A1c (HbA1c), blood glucose, and hypoglycemic events. Patient-computer interaction appears to be a valuable supplement to interaction with clinicians. Considering the need to enhance patient participation in the care of chronic illnesses, initial evidence indicates computers can play a more significant role in the future.