AI Digital Twins Assist Individuals in Managing Diabetes and Obesity

Using logged meals, the app forecasts an individual’s blood sugar reaction to those foods. It also provides tailored suggestions throughout the day, such as altering portion sizes, selecting different food combinations, or suggesting a walk post-meal. Users have the option to embrace or disregard these recommendations—perhaps broccoli isn’t their go-to vegetable, or they prefer to exercise at specific times. The app employs AI to evolve with their preferences over time. Additionally, users can communicate with human coaches for specific health inquiries.

For Buckley, Twin Health has enabled him to make healthier decisions, such as replacing frozen, prepackaged breakfast sandwiches with homemade breakfast burritos using low-carb, high-fiber wraps. He has cut out soda and now walks several miles daily.

“When I first began the program, I could hardly manage a mile before my back and knee started to ache. Now, I’m walking six and a half miles every morning,” he shares.

He appreciates receiving immediate feedback from the app and tracking his biometrics over time, noting that both his body fat percentage and blood pressure have been declining.

“That’s where I draw my motivation to keep walking and doing the work,” he explains.

Buckley has achieved his initial weight target of 300 pounds and currently weighs around 275. After decades on blood pressure medication, his doctor recently recommended a lower dosage.

When Twin Health approached the Cleveland Clinic’s health plan about implementing its program, staff endocrinologist Kevin Pantalone was initially doubtful. He opted to conduct a study himself.

“We’ve really struggled to effectively implement lifestyle changes. Patients often require multiple therapies to manage their diabetes,” he noted. “So I was definitely intrigued.” Despite the longstanding advice to simply exercise more and eat healthily, many Americans find it challenging to meet the recommended weekly physical activity levels and maintain a healthy diet.

Pantalone and his team enlisted 150 participants with type 2 diabetes, randomly assigning 100 to the Twin program and the remaining to a control group. On average, participants were 58 years old, facing obesity, and had a blood glucose level, or A1C, of 7.2 percent. A level of 6.5 percent or higher indicates diabetes. The trial aimed to determine if participants could reach an A1C of less than 6.5 percent while using fewer medications.

After a year, 71 percent of the study participants utilizing the Twin app achieved that blood sugar level with fewer medications, while only 2 percent of those in the control group did. Users of Twin also experienced greater weight loss—8.6 percent of their body weight compared to 4.6 percent in the control group.

At the start of the study, 41 percent of those using Twin were on a GLP-1 medication, but by the study’s conclusion, only 6 percent remained. In the control group, 52 percent began on a GLP-1, which increased to 63 percent by the end. The findings were published in the New England Journal of Medicine Catalyst last year.