A micro‐randomized pilot study to examine the impact of just‐in‐time nudging on after‐dinner snacking in adults with type 2 diabetes: A study protocol

Abstract

Aim: To determine whether a digital nudge soon after dinner reduces after-dinner snacking events as measured objectively by continuous glucose monitoring (CGM) in patients with type 2 diabetes (T2D).Methods: This is a single-site micro-randomized trial (MRT). People with T2D, aged 18-75 years, managed with diet or a stable dose of oral antidiabetic medications for at least 3 months, and who habitual snack after dinner at least 3 nights per week, will be recruited. Picto-graphic nudges were designed by mixed research methods. After a 2-week lead-in phase to determine eligibility and snacking behaviours by a CGM detection algorithm developed by the investigators, participants will be microrandomized daily (1:1) to a second 2-week period to either a picto-graphic nudge delivered-in-time (Intui Research) or no nudge. During lead-in and MRT phases, 24-hour glucose will be measured by CGM, sleep will be tracked by an undermattress sleep sensor, and dinner timing will be captured daily by photographing the evening meal.Results: The primary outcome is the difference in the incremental area under the CGM curve between nudging and non-nudging days during the period from 90 minutes after dinner until 04:00 AM. Secondary outcomes include the effect of baseline characteristics on treatment, and comparisons of glucose peaks and time-inrange between nudging and non-nudging days. The feasibility of 'just-in-time' messaging and nudge acceptability will be evaluated, along with the analysis of sleep quality measures and their night-to-night variability.Conclusions: This study will provide preliminary evidence of the impact of appropriately timed digital nudges on 24 -hour intertitial glucose levels resulting from altered after-dinner snacking in people with T2D. An exploratory sleep substudy will provide evidence of a bidirectional relationship between after-dinner snacking behaviour, glycaemia and sleep. Ultimately, this study will allow for the design of a future confirmatory study of the potential for digital nudging to improve health related behaviours and health outcomes.

Enabling Health Information Recommendation Using Crowdsourced Refinement in Web-Based Health Information Applications: User-Centered Design Approach and EndoZone Informatics Case Study

“…In comparison, web-based health information applications developed by trusted sources such as governments, credited health organizations, and universities provide health information with criteria that people value, such as trustworthiness, expertise, and objectivity [ 18 ]. In recent years, many of these applications have been developed globally to bypass the information recommendation algorithms of search engines and social media platforms [ 11 , 12 , 19 , 20 ]. Several applications contain mechanisms that provide personalized recommendations of nutritional information, medications, treatment plans, diagnoses or disease predictions, physical activities, or other health care services, based on users’ profiles and inputs [ 21 ].…”