Remote Monitoring Could Be Cost-Effective in Diabetes Care

Vivian Fonseca, MD


June 24, 2014

This feature requires the newest version of Flash. You can download it here.

Hello. My name is Vivian Fonseca, and I am from Tulane University in New Orleans, Louisiana. I am here at the 74th American Diabetes Association (ADA) Scientific Sessions in San Francisco. Let me tell you about a paper we presented at this meeting that is a little different from the usual studies involving diabetes.[1]

We worked on something called the "diabetes remote monitoring system." We worked with a company that develops software to promote patient adherence with appointments, medications, et cetera. We took this a little further and focused specifically on people with uncontrolled diabetes, who were either starting insulin or were already on insulin but were not controlled, and changes were being made in their treatment to try to achieve better control. Typically, with such patients, we either add a drug or try to optimize their basal insulin by titrating the dose. To follow up with these changes, we frequently call patients to make sure that they are titrating all their insulin properaly. This follow-up is labor intensive. Our staff make many phone calls, and provide a lot of diabetes education and follow-up. We wondered whether we could automate these functions.

Crucial to this is patients who are monitoring their blood glucose, which often is a challenge in terms of patient adherence. With our monitoring system, we used cellular phone technology to send the patients a message every day, asking them what their blood sugar is. Patients can enter the figure and text it back. We have used the ADA insulin adjustment algorithm to automatically make adjustments in their insulin and then assess, over time, what happened with their blood glucose control.

We randomly assigned 100 patients: one half to the monitoring system, and the other half to usual care with our nurses and other staff calling the patients to make the adjustments that are automated with the system. We were able to track these patients with an online portal that we could log on to and see who was testing their blood glucose, what was happening to blood glucose, and whether patients were making the prescribed adjustments. Patients could also enter their insulin doses.

There was a certain degree of interaction between the system and the patient, but in general, it was free of human intervention, except when the patient had a very high or very low blood sugar -- in which case the patient was automatically connected to the doctor on call in our hospital. Nobody was at risk of being managed completely by a computer.

For the first 3 months, the system performed very well. In fact, the computer advice to the patient seemed to be more effective than the usual care. However, after 3 months, some of the patients may have become tired of the system, or stopped adhering to the monitoring system. At the end of 6 months, there was no difference in the A1c levels between the usual care and the automated system groups.

We also looked at quality of life, and few differences were found. Some patients liked the system; others didn't, so it isn't for everybody. If you look at the scalability, it is possible to have hundreds of patients using the system and track them on the portal. Other capabilities of the system that were used to some extent were reminders to patients to refill medications or schedule eye or foot exams. The system also has the capability to provide feedback to the physician when a patient did not refill prescriptions, did not take prescribed medication, or did not test blood glucose. All of those capabilities are built into the system.

We can effectively use this system on a fairly large scale. Once many patients are entered into the system, it becomes very cost-effective because the cost of the system remains static. With usual care, the more patients you see, the higher the cost in terms of seeing them in clinic and telephoning them for follow-up. Automated care therefore could produce some cost savings, which is important in healthcare reform. We want patients to have better outcomes, but it has to be in a cost-effective manner.

Clearly, not all patients will like this system. Those who are technologically savvy and like to interact with a system, will like it. Some people like using their smartphones to help them make decisions. For such people, this can be a very effective tool in improving their diabetes control, which is our overall goal.

We hope to do further studies, perhaps randomizing clinics to see whether those that use the system provide better and more cost-effective care compared with clinics that don't adopt such a system. I hope to give updates at future meetings. Thank you very much for your attention.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.