Artificial Intelligence Rivals Doctors in Diabetes Prevention

A recent study from Johns Hopkins University (USA), published in the medical journal JAMA, revealed that a digital AI-based lifestyle modification program successfully reduced the risk of prediabetes with an effectiveness comparable to that of traditional programs supervised by human specialists.

This trial — the first of its kind to reach phase III clinical testing — demonstrated that digital systems can meet the standards set by the U.S. Centers for Disease Control and Prevention (CDC). Approximately 31% of participants in both groups achieved their prevention goals within one year.

The study, which included 368 participants, also showed that the AI program outperformed human-led programs in participation and completion rates: 93.4% participation compared to 82.7%, and 63.9% completion compared to 50.3%.

Researcher Benjamin Lalani noted that the biggest obstacle to the success of diabetes prevention programs is accessibility, explaining that the ease of digital access made the AI-driven application more appealing and effective for users.

Meanwhile, Nestoras Mathiodakis, the study’s principal investigator, stated that “the results prove that artificial intelligence can play a role equivalent to that of specialists in prevention programs, with a major advantage — continuous operation around the clock and freedom from logistical constraints.”

The research team believes these findings pave the way for a broader adoption of intelligent technologies in preventive healthcare, especially in communities with limited medical staff or challenges attending traditional counseling sessions.