Trial results comparing the effectiveness of fully-automated digital versus human coach-based diabetes prevention programs were posted on ClinicalTrials.gov on 2025-12-17, with a total of 368 participants enrolled.
Background
Prediabetes is a serious health condition where blood sugar levels are higher than normal but not yet high enough to be diagnosed as type 2 diabetes. It significantly increases the risk of developing type 2 diabetes, heart disease, and stroke. Lifestyle interventions, including diet and physical activity, are crucial for preventing or delaying the onset of type 2 diabetes. Diabetes Prevention Programs (DPPs) typically involve structured coaching to support these lifestyle changes. The advent of digital health solutions offers an alternative delivery method to traditional human-led coaching, raising questions about their comparative effectiveness and cost-effectiveness.
Trial design
This completed study, designated as Phase NA, enrolled 368 participants to compare two approaches for diabetes prevention. The trial focused on individuals with conditions including PreDiabetes, Hyperglycemia, Glucose, High Blood, Overweight, and Prediabetic State. The research aimed to compare the effectiveness of a fully automated digital diabetes prevention program (AI-DPP) to standard human coach-based diabetes prevention programs (Human-DPP) in promoting clinically meaningful lifestyle changes to reduce the risk of type 2 diabetes.
Key results
The trial reported several key measurements comparing the Artificial Intelligence-Based Diabetes Prevention Program (AI-DPP) and the Human Coach-Based Diabetes Prevention Program (Human-DPP):
- For the achievement of CDC's benchmark for type 2 diabetes risk reduction (binary outcome):
- The AI-DPP group had 58 participants.
- The Human-DPP group had 59 participants.
- For absolute weight change:
- The AI-DPP group showed a median change of -1.00 Kg.
- The Human-DPP group showed a median change of -1.30 Kg.
- For change in Hemoglobin A1C:
- The AI-DPP group showed a median change of 0.0 Percentage of Total Hemoglobin.
- The Human-DPP group showed a median change of -0.1 Percentage of Total Hemoglobin.
- For percentage change in weight:
- The AI-DPP group showed a median change of -1.03 percentage change.
- The Human-DPP group showed a median change of -1.43 percentage change.
- For mean weekly moderate-to-vigorous physical activity (MVPA):
- The AI-DPP group showed a median of 210.5 Minutes per Week.
- The Human-DPP group showed a median of 174.4 Minutes per Week.
- For incidence of diabetes-range A1C:
- The AI-DPP group had 8 participants.
- The Human-DPP group had 7 participants.
A primary analysis using Logistic Regression was conducted, yielding a p-value of 0.05. This analysis was performed on all randomized participants, with those not attending the 12-month visit classified as not achieving the primary composite outcome.
What this means
The posted results indicate that both fully automated digital and human coach-based diabetes prevention programs demonstrated comparable outcomes across several key metrics for individuals at risk of type 2 diabetes. The similar number of participants achieving the CDC benchmark for risk reduction, along with comparable median absolute and percentage weight changes, suggests that digital programs could offer a viable alternative to traditional human-led coaching. The AI-DPP group showed a slightly higher median weekly moderate-to-vigorous physical activity. The p-value of 0.05 from the logistic regression suggests a statistically significant difference may have been observed between the groups in the primary analysis, though the direction and specific nature of this difference would require further context.
Source
The information regarding these trial results was obtained from ClinicalTrials.gov, a public database of clinical studies. The results for the study NCT05056376, titled "Effectiveness and Cost-Effectiveness of Fully-Automated Digital vs. Human Coach-Based Diabetes Prevention Programs", were posted on 2025-12-17 on clinicaltrials.gov.