JMIR Publications

ABSTRACT

Background:

Approximately 91% of the estimated 34.2 million Americans with diabetes have type 2 diabetes. Personalized feedback, especially in response to logged blood glucose, has been identified as an effective component of mHealth apps for people with diabetes. As mHealth apps accumulate more user data, there is an opportunity for machine learning models to provide users with forecasts about their future health. One Drop’s digital program is the first to implement blood glucose forecasts that provide users with the opportunity to act on feedback about the future, and the effectiveness of those forecasts has not previously been evaluated.

Objective:

This study sought to evaluate the impact of exposure to blood glucose forecasts on blood glucose logging behavior, average blood glucose and glucose points in range.

Methods:

This retrospective cohort study examined people with type 2 diabetes who first began using One Drop to record their blood glucose between 2019-2021, some of whom received blood glucose forecasts and some of whom did not. Blood glucose measurements were logged manually, through the One Drop meter or through other synced devices. Cohorts were compared to evaluate the effect of exposure to blood glucose forecasts on logging activity, weekly average glucose, and weekly percent of glucose readings in range, controlling for potential confounding factors. Data were analyzed using ANCOVA and regression analyses.

Results:

Data from a total of 1411 One Drop members with type 2 diabetes and elevated baseline glucose (39.1% women; mean age 50.2, SD 11.8 years) were analyzed. Participants had diabetes for 7.1 years on average (SD 7.9 years). Controlling for potential confounding factors, blood glucose forecasts were associated with more frequent blood glucose monitoring, a higher percent of readings in range, and lower average blood glucose after 12 weeks. Blood glucose logging partially mediated the relationship between exposure to forecasts and average glucose.

Conclusions:

In this study, individuals who received blood glucose forecasts had significantly lower average glucose with a greater amount of glucose measurements in a healthy range. Further, these individuals checked and logged more glucose entries than those who did not receive blood glucose forecasts. Taken together, the data suggest that when administered as a part of a comprehensive mHealth program, blood glucose forecasts may significantly improve glycemic management among people living with type 2 diabetes.