Implementation of a Digital Health Intervention (CHAMP) for Self-Monitoring of Hypertension: Protocol for 3 Interlinked Implementation Studies
Bibliographic record
Abstract
BACKGROUND: Hypertension affects 31% of the global adult population. Artificial intelligence-based chatbots may support self-management of hypertension and other chronic disorders. Chronic Disease Management Program (CHAMP) is a digital health intervention designed to support chronic disease self-management, comprising a patient-facing chatbot and an artificial intelligence-augmented clinical decision support system linked to electronic medical records. OBJECTIVE: This project aims to optimize the deployment of CHAMP across primary care centers by developing implementation strategies and pilot-testing their appropriateness and effectiveness. METHODS: We report 3 interlinked studies. Study 1 is a rapid evidence review to evaluate the factors influencing the implementation of mobile health and chatbot interventions in health care settings and the strategies and processes involved. We will follow the Cochrane Rapid Reviews Methods Group methodology and use the Consolidated Framework for Implementation Research and the unified theory of acceptance and use of technology frameworks to analyze the data. Study 2 is a formative, mixed methods study to inform the state of CHAMP deployment to date, the organizational structure of primary care centers, and the barriers and facilitators influencing the implementation and scale-up of CHAMP in primary care centers. We will interview members of the CHAMP development and initial implementation team, health care providers (HCPs), primary care centers' leadership, and patient users and nonusers of CHAMP; conduct 1-day on-site visits to primary care centers; and assess the readiness to change among HCPs using validated questionnaires. Study 3 involves the development of implementation strategies for the implementation of CHAMP in primary care centers. We will develop a process map informed by the findings of studies 1 and 2 to outline the patient journey; map the barriers and facilitators influencing the implementation of CHAMP; and develop a set of implementation strategies to effectively implement CHAMP using the Expert Recommendations for Implementing Change taxonomy to define the implementation strategies and the action, actor, context, target, and time framework to define the strategy implementation processes. In parallel, we will use evidence-based behavioral science to test variations of the chatbot's messages to increase patient engagement. RESULTS: To date, the evidence review protocol (study 1) has been registered on PROSPERO. We have completed title and abstract screening and are in the full-text screening phase. For study 2, we obtained ethics approval and conducted the semistructured interviews with HCPs, primary care centers' leadership, and members of the CHAMP development and initial implementation team. We are awaiting institutional review board approval to start the interviews with patients. CONCLUSIONS: The study results will inform the further implementation and scale-up of CHAMP across primary care centers in Singapore. Successful implementation of digital health interventions to support self-management of chronic disorders may improve health care delivery without further straining health care systems. TRIAL REGISTRATION: PROSPERO CRD42024613653; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024613653. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/72942.
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 | 0.063 | 0.060 |
| Meta-epidemiology (narrow) | 0.005 | 0.004 |
| Meta-epidemiology (broad) | 0.009 | 0.010 |
| Bibliometrics | 0.005 | 0.005 |
| Science and technology studies | 0.007 | 0.004 |
| Scholarly communication | 0.005 | 0.005 |
| Open science | 0.004 | 0.005 |
| Research integrity | 0.009 | 0.010 |
| Insufficient payload (model declined to judge) | 0.068 | 0.013 |
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".