Digital Health Solutions for Type 2 Diabetes and Prediabetes: Systematic Review of Engagement Barriers, Facilitators, and Outcomes
Bibliographic record
Abstract
Background: Digital health interventions, including artificial intelligence (AI)-driven solutions, offer promise for type 2 diabetes mellitus (T2DM) and prediabetes management through enhanced self-management, adherence, and personalization. However, engagement challenges and barriers, particularly among young adults and diverse populations, persist. Existing reviews emphasize clinical outcomes while neglecting engagement factors crucial to intervention success. This review highlights engagement barriers and facilitators, offering insights into improving digital health solutions for diabetes management. Objective: The objective of this systematic literature review is to explore the barriers, facilitators, and outcomes of digital health interventions, focusing on the current state of AI applications while including partial AI and non-AI interventions, for managing and preventing T2DM and prediabetes, to inform the development of user-centered, inclusive digital health interventions for diabetes care. Unlike prior reviews, this review aims to inform the development of user-centered, inclusive digital health interventions for diabetes care, with a focus on engagement across various AI interventions and diverse populations. Methods: A systematic search of PubMed, Scopus, CINAHL, and additional sources was conducted for studies published between January 2016 and October 2025. Eligibility criteria included English-language, peer-reviewed studies focused on digital health interventions for adults with T2DM or prediabetes, reporting engagement, barriers, facilitators, or outcomes. Data were synthesized narratively using thematic analysis, guided by self-determination theory and user-centered design. Quality appraisal was conducted using Critical Appraisal Skills Program, Mixed Methods Appraisal Tool, and AMSTAR-2 tools. Results: From the 37 studies (14 quantitative, 3 qualitative, 7 mixed-methods, and 13 reviews), interventions comprised 19 AI-driven (eg, chatbots, ML models, and conversational agent or hybrid), 3 partially AI-driven, and 15 non-AI solutions (eg, apps and lifestyle programs), mostly from the USA (n=15). Key barriers to engagement included inadequate personalization (15/37, 41%), environmental constraints (11/37, 11%), cultural and language mismatches (14/37, 38%), and AI-specific concerns (eg, bias and privacy). Facilitators included personalized feedback (19/37, 51%), cultural tailoring (17/37, 46%), user-friendly design, and peer support. AI-driven interventions demonstrated moderate improvements in clinical outcomes (eg, lowering HbA1c, weight loss, and normoglycemia conversion). However, these tools often struggled with keeping users involved and building trust. Non-AI solutions performed similarly but lacked adaptive features. Conclusions: This review offers novel insights by synthesizing engagement barriers and facilitators across AI and non-AI intervention domains, often neglected in previous studies. It highlights the necessity for testing adaptive, culturally tailored, and user-centered AI interventions to address engagement challenges in T2DM and prediabetes management. Integrating personalization, precision, and value-based care can improve outcomes and scalability. The findings guide the creation of inclusive, AI-driven solutions aligned with self-determination theory and user-centered design principles.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".