JMIR Publications

ABSTRACT

Background:

For adolescents living with type 1 diabetes (T1D), completion of multiple daily self-management tasks, such as monitoring blood glucose and administering insulin, are challenging due to psychosocial and contextual barriers. Those barriers are hard to assess accurately and specifically using traditional retrospective recall. Ecological momentary assessment (EMA) uses mobile technologies to assess the contexts, subjective experiences, and psychosocial processes that surround self-management decision-making in daily life. However, the rich data generated via EMA has not been frequently examined in T1D or integrated with machine learning analytic approaches.

Objective:

The goal of this study was to develop a machine-learning algorithm to predict risk for missed self-management of young adults with T1D diabetes. To achieve this goal, we trained and compared a number of machine learning models through a learned filtering architecture (LFA) to explore the extent to which EMA data was associated with completion of two self-management behaviors: mealtime self-monitoring of blood glucose (SMBG) and insulin administration.

Methods:

We analyzed data from a randomized controlled pilot study using machine learning-based filtering architecture to investigate whether novel information related to contextual, psychosocial, and time-related factors (i.e., time of day) relate to self-management. We combined EMA-collected contextual and insulin variables via the MyDay mobile app with Bluetooth blood glucose data to construct machine learning classifiers that predicted the two self-management behaviors of interest.

Results:

With 1,231 day-level SMBG frequency counts for 45 participants, demographic variables and time-related variables were able to predict whether daily SMBG was below the clinical threshold of 4 times a day. Using the 1,869 data points derived from 31 participants’ app-based EMA data, our LFA method was able to infer non-adherence events with a high accuracy and precision. Although the recall score is low, there is high confidence that the non-adherence events identified by the model are truly non-adherence.

Conclusions:

Combining EMA data with machine learning methods showed promise in the relationship to risk for non-adherence. Next steps include collecting larger datasets that would more effectively power a classifier that can be deployed to infer individual behavior. Improvements in individual self-management insights, behavioral risk predictions, enhanced clinical decision-making and just-in-time patient support in diabetes could result from this type of approach.