Diabetes mellitus is a costly, chronic, metabolic disorder that is commonly undiagnosed and frequently inadequately treated. The Diabetes Control and Complications Trail (DCCT) in 1993 and the U.K. Prospective Diabetes Study (UKPDS) in 1998 demonstrated clearly that intensive control of blood glucose has a dramatic effect on reducing the incidence of complications, improving the quality and length of life and reducing costs of care for these patients. Despite that, numerous studies have shown a large gap between current practice and bestpractice care for patients with diabetes. Recently, disease management has been suggested as an innovative strategy to improve the quality of care delivered to patients with chronic diseases. A diabetes disease management program was implemented in the University of Pennsylvania Health System (UPHS), an academic-based, integrated healthcare delivery system, in the spring of 1997. Prior to implementation of the program, a snapshot assessment of current practice was obtained by random retrospective review of approximately 40 charts from patients with diabetes. We now report data from 1779 patients enrolled in the program. Just over half the patients are female and the mean age is 60.2 years. A highly significant improvement has been noted in hemoglobin A1c (HbA1c) levels. Before enrollment, from the snapshot review, average HbA1c values were 9.03%±2.11% and this has fallen in study patients to 8.30%±2.06% (p = 0.03). If only study patients are evaluated, the mean preprogram HbA1c value was 8.31%±2.10% and, in those cases where repeat values are available, it has fallen to 8.07%±1.62% (p =.07). Overall, by the time of this report, 60% of the patients enrolled in the disease management program have a HbA1c level of less than 8%, considerably higher than the 40% documented before the program was started. Impressive improvements also were noted in several key process measures, including foot examinations, blood sugar measurement, and blood pressure documentation. While these are encouraging early outcomes, we consider it a work-in-progress as we continually modify and enhance program components and processes of care delivery to constantly improve the outcomes. It is also important to emphasize that our results demonstrate clearly that this approach is entirely appropriate and can be sucessfully implemented in a large, complex academic health system such as the UPHS. Medical students and residents are now intensely involved at all stages of the program and have revealed a high level of enthusiasm and excitement as they find themselves being part of this dramatic new healthcare re-engineering process.