Health recommender systems to facilitate collaborative decision-making in chronic disease management: A scoping review
Bibliographic record
Abstract
Objective: Health recommender systems (HRSs) are increasingly used to complement existing clinical decision-making processes, but their use for chronic diseases remains underexplored. Recognizing the importance of collaborative decision making (CDM) and patient engagement in chronic disease treatment, this review explored how HRSs support patients in managing their illness. Methods: A scoping review was conducted using the framework proposed by Arksey and O'Malley, advanced by Levac et al., in line with the PRISMA-ScR checklist. Quantitative (descriptive numerical summary) and qualitative (inductive content analysis) methods wered used to synthesize the data. Results: Forty-five articles were included in the final review, most commonly covering diabetes (9/45, 20%), mental health (9/45, 20.0%), and tobacco dependence (7/45, 15.6%). Behavior change theories (10/45, 22.2%) and authoritative sources (10/45, 22.2%) were the most commonly referenced sources for design and development work. From the thematic analysis, we conclude: (a) the main goal of HRSs is to induce behavior change, but limited research investigates their effectiveness in achieving this aim; (b) studies acknowledge that theories, models, frameworks, and/or guidelines help design HRSs to elicit specific behavior change, but they do not implement them; (c) connections between CDM and HRS purpose should be more explicit; and (d) HRSs can often offer other self-management services, such as progress tracking and chatbots. Conclusions: We recommend a greater emphasis on evaluation outcomes beyond algorithmic performance to determine HRS effectiveness and the creation of an evidence-driven, methodological approach to creating HRSs to optimize their use in enhancing patient care. Lay summary: Our work aims to provide a summary of the current landscape of health recommender system (HRS) use for chronic disease management. HRSs are digital tools designed to help people manage their health by providing personalized recommendations based on their health history, behaviors, and preferences, enabling them to make more informed health decisions. Given the increased use of these tools for personalized care, and especially with advancements in generative artificial intelligence, understanding the current methods and evaluation processes used is integral to optimizing their effectiveness. Our findings show that HRSs are most used for diabetes, mental health, and tobacco dependence, but only a small percentage of publications directly reference and/or use relevant frameworks to help guide their design and evaluation processes. Furthermore, the goal for most of these HRSs is to induce behavior change, but there is limited research investigating how effective they are in accomplishing this. Given these findings, we recommend that evaluations shift their focus from algorithms to more holistic approaches and to be more intentional about the processes used when designing the tool to support an evidence-driven approach and ultimately create more effective and useful HRSs for chronic disease management.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.021 | 0.072 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.005 |
| Bibliometrics | 0.010 | 0.011 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.003 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".