The Service of Research Analytics to Optimize Digital Health Evidence Generation: Multilevel Case Study (Preprint)
Bibliographic record
Abstract
BACKGROUND The widespread adoption of digital health interventions for chronic disease self-management has catalyzed a paradigm shift in the selection of methodologies used to evidence them. Recently, the application of digital health research analytics has emerged as an efficient approach to evaluate these data-rich interventions. However, there is a growing mismatch between the promising evidence base emerging from analytics mediated trials and the complexity of introducing these novel research methods into evaluative practice. OBJECTIVE This study aimed to generate transferable insights into the process of implementing research analytics to evaluate digital health interventions. We sought to answer the following two research questions: (1) how should the service of research analytics be designed to optimize digital health evidence generation? and (2) what are the challenges and opportunities to scale, spread, and sustain this service in evaluative practice? METHODS We conducted a qualitative multilevel embedded single case study of implementing research analytics in evaluative practice that comprised a review of the policy and regulatory climate in Ontario (macro level), a field study of introducing a digital health analytics platform into evaluative practice (meso level), and interviews with digital health innovators on their perceptions of analytics and evaluation (microlevel). RESULTS The practice of research analytics is an efficient and effective means of supporting digital health evidence generation. The introduction of a research analytics platform to evaluate effective engagement with digital health interventions into a busy research lab was ultimately accepted by research staff, became routinized in their evaluative practice, and optimized their existing mechanisms of log data analysis and interpretation. The capacity for research analytics to optimize digital health evaluations is highest when there is (1) a collaborative working relationship between research client and analytics service provider, (2) a data-driven research agenda, (3) a robust data infrastructure with clear documentation of analytic tags, (4) in-house software development expertise, and (5) a collective tolerance for methodological change. CONCLUSIONS Scientific methods and practices that can facilitate the agile trials needed to iterate and improve digital health interventions warrant continued implementation. The service of research analytics may help to accelerate the pace of digital health evidence generation and build a data-rich research infrastructure that enables continuous learning and evaluation.
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 | 0.060 | 0.097 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.010 | 0.007 |
| Scholarly communication | 0.008 | 0.006 |
| Open science | 0.003 | 0.010 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.008 | 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".