VUMC team forms new ways to track, manage diabetes
A VUMC team is developing new ways to track and manage diabetes among the adult primary care population.
As a whole, the Adult Primary Care Center (APCC) currently follows some 2,000 patients with diabetes, or 80 diabetics per doctor, and management of this population is becoming a decidedly collaborative affair.
Diabetes is the leading cause of blindness and end-stage renal disease and a risk factor for heart attack and stroke. Five percent of adults are known to have diabetes and another five percent have a mild, incipient, unrecognized form of the disease. In 1997, 12 percent of patients admitted to Vanderbilt University Hospital had diabetes.
National standards for management of diabetes call for testing each patient’s average blood sugar level every six months. It’s called HbA1c testing and results serve as a firm indicator of whether to look for acute problems in the future. Formerly, doctors couldn’t review these test results without searching patient records one at a time.
APCC physicians now receive concise quarterly reports on their patients with diabetes, indicating the date and result of each of a patient’s most recent HbA1c test. The report is from Dr. Alan Graber, professor of Medicine and clinical director for Endocrinology and Diabetes.
A year ago, a Vanderbilt diabetes team began swapping strategy with other diabetes teams from around the country. Vanderbilt has been among some 33 institutions — such as Yale, Mayo Clinic and the University of Washington — participating in the Chronic Conditions Breakthrough Series, sponsored by the Institute for Healthcare Improvement (IHI). The series is focused on two populations: diabetics and the frail elderly.
The Chronic Conditions Breakthrough Series concludes this month with a national meeting. Vanderbilt’s core team includes Graber; Dr. Tom A. Elasy, assistant professor of Medicine; Doris Quinn, Ph.D., Nursing Systems Coordinator; and Joyce B. Thomas of the Center for Clinical Improvement.
“Chronic illness care is under-emphasized in many medical organizations,” Graber said. “We were already thinking about making this effort with diabetes, but IHI helped structure it.”
The Vanderbilt team began by setting some goals. Because the HbA1c result is such a firm indicator of things to come, the test is a dependable surrogate for outcomes data such as admission rates, morbidity and mortality. The team’s initial goals were that, at any given time, 90 percent of patients with diabetes seen by the APCC were to have an HbA1c test performed within the previous six months, 80 percent were to have an HbA1c value under 10 percent, and 40 percent were to have a value under 8 percent.
The difference between values of 10 percent and 8 percent is significant in terms of lowering risk of eye and kidney complications.
At the start, IHI laid out a model process for improvement based on the PDSA cycle — plan, do, study, act. The model also identified four key aspects of the health system: support for patient self-management, care delivery system design, clinical decision support and clinical information systems.
“The model may sound quite basic and generic,” Graber said, “but we’ve found it to be very useful.”
With goals defined, the team set about creating and refining a model whereby Dr. Graber and his team of nurse educators, dieticians and a data base manager assist primary care doctors in achieving better outcomes in their patients with diabetes.
The collaborative effort provides guidelines for PCPs both on when to refer patients to an endocrinologist and when to refer patients for diabetes self management education provided by the Diabetes Improvement Program (DIP) supervised by Graber.
An electronic patient registry is the foundation for diabetes collaboration at Vanderbilt. The registry was initially populated by querying certain VUMC systems (Epic and DB2) for the diagnosis codes that indicate diabetes. With help from Dario Giuse, Ph.D., associate professor of Biomedical Informatics, the team learned how to run this registry periodically against the lab system.
The team is encouraged by the results of collaboration and is revising goals upward. From the start, 84 percent of diabetics followed by the APCC (the goal was 80 percent) were found to have an HbA1c value under 10 percent, and 50 percent of these patients (the goal was 40 percent) were found to have a value under 8 percent.
“As a result, we’re saying the original goals were too modest,” Graber said.
“We now think that 80 percent should have an HbA1c below 8 percent.”
For more information contact Graber by email or at 3-5945.