Chronic Care Model

A way of delivering health care to groups of people with chronic diseases, including diabetes. The current model for most health-care delivery emphasizes acute care, dealing with individual problems as they arise (such as when a person goes to the doctor for a cold or the flu). However, that approach to health care is not the best for treating diabetes.

Clinical trials have shown time and again that carefully controlling blood glucose levels and blood pressure can dramatically lower a person’s risk of developing diabetic complications, but most people with diabetes are not getting adequate help in doing this. Diabetes care in the United States is compromised by a number of things, including time constraints on health professionals, lack of coordination among care providers, and limited access to diabetes educators and physicians who specialize in diabetes.


Health-care providers and researchers at the University of Pittsburgh Medical Center (UPMC) in Pittsburgh, Pennsylvania, have designed a chronic care model for treating diabetes that stresses prevention. It emphasizes diabetes education, which was shifted from hospitals to primary-care physician offices to make it more easily available. The model gives primary-care physicians a variety of computerized data: They can view overall trends among their patients in HbA1c (glycosylated hemoglobin, a measure of blood glucose control), LDL (“bad”) cholesterol, and urine microalbumin (a measure of kidney function) levels over a given period of time, or they can view the trends of a single patient. They are also given information on diabetes-related outcomes in their practice compared to outcomes across the UPMC system. Finally, the model emphasizes following up-to-date management guidelines based on scientific evidence, for both doctors and diabetes educators.

In a study reported in the journal Diabetes Care in 2006, researchers at UPMC compared the effects of their chronic care model with those of two other models of health-care delivery, “provider education only” and “usual care,” when each was applied to underserved people with diabetes. There was a marked decline in HbA1c levels (signifying improved blood glucose control) in the group of people assigned to the chronic care model, but not in the other two groups. Those in the chronic-care-model group had lower LDL cholesterol levels and higher rates of blood glucose self-monitoring than the other groups. Furthermore, within the chronic-care-model group, improvements were seen in HDL (“good”) cholesterol levels, diabetes knowledge, and “empowerment scores” (how much control people felt they had over their health).