We have previously introduced Micro-Randomised Trials (MRTs) in an earlier post. If you’re interested in learning more about what MRTs are and how they work, please refer to that post before reading further. MRTs have emerged as a foundational methodology for developing Just-In-Time Adaptive Interventions (JITAIs), which aim to deliver the right support to the right person at the right time (Nahum-Shani et al., 2018; Klasnja et al., 2015).
Diabetes management is one of the most promising areas for MRT implementation because individuals make dozens of health-related decisions every day that influence glycaemic control.
Modern diabetes care increasingly incorporates technologies such as Continuous Glucose Monitors (CGMs), smartwatches and activity trackers, smartphone applications, connected insulin devices, artificial intelligence (AI) and predictive analytics. These technologies create opportunities to provide personalised support throughout the day while simultaneously generating large volumes of behavioural and physiological data (Riley et al., 2011).
Physical Activity Prompts
Physical activity is an important component of diabetes self-management and has been shown to improve glycaemic control, insulin sensitivity, and cardiometabolic health (Colberg et al., 2016).
Using an MRT design, a diabetes management system could detect prolonged sitting and randomly deliver prompts encouraging users to take a short walk or movement break. Researchers can then determine whether physical activity increases immediately after the prompt, whether glucose excursions are reduced, which times of day generate the strongest response, and whether intervention effectiveness varies across individuals.
Dietary Decision Support
Nutrition decisions are among the most frequent and influential behaviours affecting glucose control. An MRT could evaluate the effectiveness of different dietary interventions delivered before meals, or following episodes of elevated glucose levels. For example, participants might receive meal planning suggestions, lower-glycaemic food alternatives, portion-control reminders, or real-time feedback linked to CGM readings. Researchers can then identify which approaches produce the greatest improvements in dietary behaviour and postprandial glucose responses.
Medication Adherence Support
Medication adherence remains a major challenge in diabetes management, particularly among individuals requiring multiple medications. MRTs can be used to evaluate the effectiveness of reminder systems delivered at medication-taking opportunities. These reminders may vary by timing, format, content, or delivery channel. Such studies can help identify the most effective strategies for improving adherence while minimising notification burden and alert fatigue.
Stress and Wellbeing Interventions
Psychological stress influences both glucose regulation and diabetes self-management behaviours (Lloyd et al., 2018). Using wearable devices and smartphone sensors, future interventions may detect indicators of stress and deliver adaptive support such as mindfulness exercises, breathing techniques, motivational messages, or behavioural coping strategies.
MRTs provide a rigorous framework for evaluating the short-term effectiveness of these interventions under real-world conditions.
From Micro-Randomised Trials to Just-In-Time Adaptive Interventions
The ultimate goal of MRT research is not simply to evaluate intervention effectiveness but to optimise interventions before large-scale implementation. Data generated through MRTs can be used to build Just-In-Time Adaptive Interventions (JITAIs), which continuously tailor support based on an individual’s context, behaviour, and physiological state (Nahum-Shani et al., 2018).
In diabetes management, future JITAIs may integrate real-time CGM data, physical activity patterns, sleep behaviours, dietary information, medication adherence data, and environmental and contextual factors. These systems could determine not only whether an intervention should be delivered, but also what type of intervention should be delivered and when it is most likely to be effective. AI and machine learning techniques may further enhance these systems by identifying complex patterns and adapting intervention strategies over time.
The Future of Diabetes Care
The emerging model of diabetes care is shifting from periodic self-monitoring toward adaptive, AI-supported co-management. Instead of relying solely on retrospective review of glucose readings, future systems will continuously monitor behavioural and physiological data and provide personalised support in real time. MRTs play a critical role in this transition by providing the scientific foundation needed to optimise adaptive interventions before implementation at scale. By helping researchers understand what works, for whom, when, and under what circumstances, MRTs can accelerate the development of more effective digital health solutions that support people living with diabetes in their daily lives.
As healthcare increasingly embraces wearable technologies, AI, and precision health approaches, MRTs are likely to become one of the most important methodologies for developing the next generation of personalised diabetes interventions.
At ReDAdvise, we are passionate about applying innovative methodologies including MRTs and Agentic AI to design, evaluate, and optimise interventions that improve outcomes for individuals living with chronic conditions.
References
- Klasnja P, Hekler EB, Shiffman S, Boruvka A, Almirall D, Tewari A, Murphy SA. Micro-randomized trials: An experimental design for developing just-in-time adaptive interventions. Health Psychology. 2015;34(S):1220–1228.
- Nahum‐Shani I, Smith SN, Spring BJ, Collins LM, Witkiewitz K, Tewari A, Murphy SA. Just-in-Time Adaptive Interventions (JITAIs) in mobile health: Key components and design principles for ongoing health behavior support. Annals of Behavioral Medicine. 2018;52(6):446–462.
- Riley WT, Rivera DE, Atienza AA, Nilsen W, Allison SM, Mermelstein R. Health behavior models in the age of mobile interventions: Are our theories up to the task? Translational Behavioral Medicine. 2011;1(1):53–71.
- Colberg SR, Sigal RJ, Yardley JE, Riddell MC, Dunstan DW, Dempsey PC, et al. Physical activity/exercise and diabetes: A position statement of the American Diabetes Association. Diabetes Care. 2016;39(11):2065–2079.
- Murphy SA. An experimental design for the development of adaptive treatment strategies. Statistics in Medicine. 2005;24(10):1455–1481.