‘Smart speaker’ shows potential for better self-management of type 2 diabetes

Type 2 diabetes patients have access to an assortment of digital health tools, including iPhone mobile apps and remote patient monitoring devices, all of which help them manage their blood sugar levels. They can take the test at home, usually with a glucometer, and then report that data to their healthcare team through an app or web portal, by calling on the phone, or while waiting for an appointment.

But these monitoring tools assume a level of comfort with technology and sometimes math skills that many patients don’t have. And they all need something that is lacking: clinicians.

Now, researchers at Stanford Medicine have created a voice-based artificial intelligence application that runs on a device already familiar to tens of millions of Americans: a “smart speaker,” commonly used to listen to music and check the weather. The app tells patients the right dose of insulin without needing to contact their doctor’s office or wait for an appointment.

“People just don’t have a lot of access to care,” said Ashwin Nayak, MD, assistant professor of clinical medicine. “We want to empower patients to do it themselves.” Nayak is the first author of the paper based on the study, published in Open JAMA Network.

Participants in a randomized trial who used the system achieved optimal blood sugar levels significantly faster than the control group; they also preferred to take the prescribed amount of insulin at the time they were supposed to.

Additionally, they reported lower levels of emotional distress related to their diabetes.

Most patients with type 2 diabetes do not benefit from continuous glucose monitors or insulin pumps commonly used by patients with type 1 diabetes. Insulin pumps provide a constant amount of insulin to make up the deficit in insulin, in which cells that should produce insulin fail to do so. In contrast, type 2 diabetics have insulin resistance, so their insulin needs tend to be less constant.

Small study shows results

The study followed 32 people with type 2 diabetes who were all taking insulin and struggling to achieve healthy blood sugar levels. Half of them received a speaker preloaded with voice artificial intelligence software created by the research team.

Each participant’s insulin protocol, including initial insulin dose, fasting blood glucose goal, and insulin dose instructions, was included in their device software.

Participants assigned to the AI ​​group were asked to check-in each day using the phrase “Check in with a clinical trial,” which triggered a conversation in which the participant reported clinical data, such as recent consumption insulin and fasting blood sugar reading.

At the end of the conversation, the software provided updated instructions by responding, for example, “OK, continue doing what you are doing” or “Increase your dose” by a specified amount. The platform goes beyond remote monitoring by enabling remote intervention on patients for the first time, the researchers said.

Participants randomized to the standard care group also received a smart speaker, configured with daily reminders to complete their diary, but not containing the app. These patients monitored their blood sugar as usual and reported the data to their providers online or by phone. If they needed a dose change, their doctor contacted them.

Although the trial sample size was small, the impact was dramatic. During the eight-week study period, 81% of patients in the group who used the app achieved glycemic control of their diabetes, compared to 25% of patients who received traditional care. Patients in the AI-run trial had their insulin dose adjusted more frequently and needed significantly fewer doctor appointments to get their diabetes under control.

According to the researchers, this voice-activated device has great potential to improve access, usability and convenience, especially for older patients with type 2 diabetes.

“Patients don’t have the tools to navigate the system and manage the complexity of their treatments,” said Kevin Schulman, MD, professor of medicine and co-director of the Clinical Excellence Research Center at Stanford, who led the study. “If we want to meet people where they are, we’ll need technology to help us. With this approach, AI and voice interface come together to try to solve a huge problem.”

The system used in the trial was programmed in English, but it can easily be adjusted for speakers of other languages, the researchers said. The app can also be programmed to monitor the patient’s response to other diabetes medications they are taking, in addition to insulin. Researchers predict it could also be used to monitor other chronic diseases.