Dr. Ben Littenberg (MD, FACP) is the principal investigator for the study. He is the Henry and Carleen Tufo Professor of Medicine, Professor of Nursing and Director of General Internal Medicine for the University of Vermont College of Medicine. He is also CEO of Vermont Clinical Decision Support which distributes the Vermedx® Diabetes Information System (VDIS), the subject of this study.
Dr. Littenberg sat down with Healthcare Technology News to talk about his findings.
HTN: Can you please talk about Wagner's Chronic Care Model and its influence on your work?
Ben Littenberg: The Chronic Care Model is a very influential idea about how to do a good job in taking care of people with ongoing medical problems like high blood pressure, arthritis, and in our case, diabetes. It tells what resources the health care system must have to get the most out of the interaction between the doctor and the patient. These include systems to support clinical decision making, up-to-date and easy-to-use clinical information, reminders, support for patient self-management, and so on. Turning a typical primary care medical office into a fully-functional chronic care model practice is a good thing, but it is complex, expensive and time-consuming. It requires extensive retraining of the physicians, nurses and other staff, installation of comprehensive electronic medical records systems, changes in the way patients and staff interact, and even an overhaul of the way care is paid for. So, very few practices have successfully transformed themselves. When we designed Vermedx(R), we took as much of the chronic care model as we could while still maintaining very low cost per patient, very little disruption of the clinical work-flow, and no requirement that the practice invest money on computers or other fixed costs.
HTN: What is the current state of technology to assist in the management of diabetes?
Ben Littenberg: Drugs for diabetes have gotten much better in the last 20 years. Also, we know a lot now about how to prevent the complications of diabetes - heart attacks, strokes, kidney failure, blindness, amputations, and so on. However, it takes a lot of very careful attention to lots and lots of data. How high is the blood sugar? When was it measured? When should the cholesterol be tested again? How close is the patient to goal? Office systems have failed to keep up with the needs of the patients and providers. Paper records just don't work well in this setting. A typical laboratory report shows the numeric value of the result and a so-called "normal" range, but it doesn't tell you what the test means or what to do about it.
Most offices have no organized method for keeping track of all the blood and urine tests used in diabetes, let alone a system that reminds the patient when a test is due and even advises the doctor on what to do.
HTN: How is VDIS different?
Ben Littenberg: Vermedx automatically collects the laboratory test results on diabetic patients and reports them to the doctor (or other health care provider). It puts the test results into context, shows trends over time, and suggests actions. It also takes action when a test should have been done, but wasn't. It sends reminders to the practice and letters (from the providers) to the patients encouraging them to get into care. If the test results are very high, it also notifies the patient to get back to the practice to change the care plan. Vermedx also provides a population view of the doctor's entire panel of diabetic patients: who is doing well, who needs help, and how the practice is doing compared to others.
HTN: Your study found a sustained continuing decrease in utilization that is striking. Please tell us about the results.
Ben Littenberg: With NIH funding, we conducted a long-term randomized study of VDIS in community primary care. Over 7,000 patients in 64 practices participated. The patients in that study who received VDIS reported far fewer trips to the Emergency Room, unscheduled hospital admissions and doctor visits than patients who didn't get VDIS. The problem was that we didn't have access to the medical bills in that study. The recent study in the American Journal of Managed Care looked at patients getting Vermedx as part of their usual care and was able to compare the insurance claims paid to other diabetic adults who did not have Vermedx. Before VDIS was available, both groups had steadily rising costs month-by-month. The control patients' claims kept going up, but less than a year after turning on the system, the VDIS patients' claims started going down. The savings were $504 per patient in the first year, went up to $3,563 in year 4, and kept getting better!
HTN: For the VDIS patients, what were the differences in health outcomes?
Ben Littenberg: VDIS patients are much more likely to get their laboratory tests on time. Although they are no more likely to reach goals for blood sugar or cholesterol, they spend less time in the Emergency Room and hospital, which indicates a substantial improvement in quality of life.
HTN: Understanding that your study does not suggest causality between VDIS and utilization, what is your hypothesis about the relationship?
Ben Littenberg: Actually, the first study was randomized and therefore, does provide good evidence of causality. However, nobody is entirely certain how it works. Since the blood test results aren't different, it probably doesn't work by preventing complications. One theory is that improved communications give patients confidence - they know that they will be back in touch with the doctor and don't need to call for every little issue. It's like those reminder postcards to get your car's oil changed. You know you're going to bring the car in soon, so you might be less anxious about getting service right away for some funny noise under the dashboard. Of course, you're also being seen on a schedule - before you have an emergency!
HTN: VDIS sends alerts to both physicians and patients. What is your view of the relative effectiveness of focus on the physician or focus on patient self-management?
Ben Littenberg: They both seem to be valuable. The core idea behind the Chronic Care Model is to activate the patient to get involved in their care AND prepare the practice to meet that patient's needs. Working on both ends of the relationship is much more effective than either one alone.