Ecological momentary assessment in adolescents with type 1 diabetes: A mixed-methods feasibility study.

Abstract

Aims: Diabetes self-management is dynamic and subject to the influence of contextual variables. Ecological momentary assessment (EMA) facilitates the examination of temporal processes and the influence of contextual variables but is underutilised in paediatric type 1 diabetes research. This study examined the feasibility of using text-message-based EMA and how patient and contextual factors predict adherence to EMA procedures.

Methods: Adolescents (12-18 years old) with type 1 diabetes (n = 66) watched an EMA training video, completed a 5-question EMA quiz, and then participated in a 3-month EMA study. Mixed-effects models and paired t-tests examined predictors (i.e. HbA1c, participant age, EMA quiz score, study time-point, time of day, day of week) of response rate and latency. Eighteen participants, purposively sampled based on response rates, were interviewed, and thematic analysis was used to summarize themes related to their EMA experience.

Results: Average response rate was 70% and average response latency was 77.83 min. Response rates were higher on weekdays. Response latency was longer for morning and weekday prompts. In mixed-effects models, response rates decreased over time and were lower for youth with higher HbA1c. Better quiz scores predicted higher response rates and shorter response latency. Qualitative data revealed themes highlighting (1) methodological considerations for research using EMA and (2) how EMA positively affected participants.

Conclusions: Text-message-based EMA is a feasible method for capturing near-real-time patient experiences related to diabetes management. Combined with diabetes technology, EMA has significant potential for improving our understanding of the behavioural aspects of diabetes self-management.