Development of an Evaluation Index System for Health Recommender Systems Based on the Health Technology Assessment Framework: Cross-Sectional Delphi Study
Bibliographic record
Abstract
Background: Health recommender systems (HRSs) are digital platforms designed to deliver personalized health information, resources, and interventions tailored to users' specific needs. However, existing evaluations of HRSs largely focus on algorithmic performance, with limited scientific evidence supporting user-centered assessment approaches and insufficiently defined evaluation metrics. Moreover, no unified or scientifically validated framework currently exists for evaluating these systems, resulting in limited cross-study comparability and constraining regulatory and implementation decision-making. Objective: This study aimed to develop a comprehensive, consensus-based evaluation index system for HRSs grounded in the health technology assessment (HTA) framework. Methods: This cross-sectional study used a 2-round Delphi process conducted with 18 experts comprising clinicians, digital health researchers, and policymakers who possessed relevant professional experience and domain knowledge in HRSs. The age range of the experts was between 30 and 58 years, with 67% (n=12) of them possessing over 10 years of professional experience. On the basis of literature analysis and HTA principles, a preliminary indicator set comprising 5 primary and 16 secondary indicators was constructed. Experts rated the importance of each indicator using a 5-point Likert scale and provided qualitative suggestions for refinement. After the Delphi process, the analytic hierarchy process was applied to determine indicator weights and assess consistency. Results: The Delphi survey reached full participation in the first round (18/18, 100%) and maintained an 88.9% (16/18) response rate in the second round. The final evaluation index system of HRSs contained 5 first-level indicators (performance, effectiveness, safety, economy, and social appropriateness) and 18 second-level indicators. The mean importance scores of the second-level indicators ranged from 4.25 (SD 0.45) to 5.00 (SD 0.00), with coefficients of variation between 0.000 and 0.220. Among the first-level indicators, safety received the highest weight (0.289), followed by social appropriateness (0.251), effectiveness (0.193), performance (0.136), and economy (0.132). Conclusions: This study presents an evaluation index system for HRSs grounded in the HTA framework and validated through expert consensus. The resulting framework not only provides actionable guidance for the design, optimization, and implementation of HRSs but also fills a methodological gap in the field by offering quantifiable, hierarchical evaluation indicators with validated weighting. Future research will involve iterative refinement and empirical validation of the system in real-world deployment settings, thereby enabling continuous improvement and facilitating the establishment of unified evaluation standards for HRS research and practice.
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 | 0.198 | 0.149 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.006 | 0.003 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.002 | 0.007 |
| Research integrity | 0.002 | 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".