Search for a command to run...
Monitoring of glucose plays an essential role in the management of diabetes. However, to fully understand and meaningfully interpret glucose levels, additional information on context is necessary. Important contextual factors include data on behaviours such as eating, exercise, medication-taking and sleep, as well as data on mental health aspects such as stress, affect, diabetes distress and depressive symptoms. This narrative review provides an overview of the current state and future directions of precision monitoring in diabetes. Precision monitoring of glucose has made great progress over the last 5 years with the emergence of continuous glucose monitoring (CGM), automated analysis of new glucose variables and visualisation of CGM data via the ambulatory glucose profile. Interestingly, there has been little progress in the identification of subgroups of people with diabetes based on their glycaemic profile. The integration of behavioural and mental health data could enrich such identification of subgroups to stimulate precision medicine. There are a handful of studies that have used innovative methodology such as ecological momentary assessment to monitor behaviour and mental health in people's everyday life. These studies indicate the importance of the interplay between behaviour, mental health and glucose. However, automated integration and intelligent interpretation of these data sources are currently not available. Automated integration of behaviour, mental health and glucose could lead to the identification of certain subgroups that, for example, show a strong association between mental health and glucose in contrast to subgroups that show independence of mental health and glucose. This could inform precision diagnostics and precision therapeutics. We identified just-in-time adaptive interventions as a potential means by which precision monitoring could lead to precision therapeutics. Just-in-time adaptive interventions consist of micro-interventions that are triggered in people's everyday lives when a certain problem is identified using monitored behaviour, mental health and glucose variables. Thus, these micro-interventions are responsive to real-life circumstances and are adaptive to the specific needs of an individual with diabetes. We conclude that, with current developments in big data analysis, there is a huge potential for precision monitoring in diabetes.