Factors Affecting the Receptiveness of Chinese Internists and Surgeons Toward Artificial Intelligence–Driven Drug Prescription: Protocol for a Systematic Survey Study
Bibliographic record
Abstract
BACKGROUND: Recently, we developed and tested an autonomous artificial intelligence (AI) agent for prescribing a drug to prevent severe acute graft-versus-host disease in patients receiving human leukocyte antigen haplotype-mismatched hematopoietic cell transplants in a prospective clinical trial. Our experience in this proof-of-concept study suggests that physicians and patients can be receptive to autonomous AI prescription. However, the generalizability of our conclusion requires testing in additional clinical settings. Before broadening the scope of study of AI-driven drug prescriptions, it is important to quantify the factors that influence a physician's receptiveness to AI prescription. OBJECTIVE: We aim to systematically interrogate physicians' receptiveness to AI prescription in China. METHODS: We have designed a research protocol to survey a diverse range of factors that may affect physicians' receptiveness to AI prescription systems, including the physicians' personal attributes and their perceptions of the importance of various technological, institutional, and governmental attributes. The survey will be conducted in 2 phases. In phase 1, the survey will be limited to the Tianjin metropolitan area, enlisting >250 physicians from approximately 2 tier-1, 3 tier-2, and 3 tier-3 hospitals. In phase 2, we will survey metropolitan areas in ≥10 additional province-level administrative divisions, enlisting >1250 additional physicians from >15 tier-1, >15 tier-2, and >15 tier-3 hospitals. We hypothesize that physicians can be broadly classified into distinct psychological profile types, and furthermore, that these types are plausibly mediated by the locales where the physicians are employed and the physicians' demographics, educational and job experience, clinical subspecialties, and previous knowledge of and experience with AI. Clustering methods, including t-distributed stochastic neighbor embedding and hierarchical clustering, will be performed on respondent data to identify the distinct psychological profile types of the physicians. Multiple-variable regression and mediation analyses will be conducted to identify potential underlying mechanisms mediating physicians' receptiveness to AI prescription. RESULTS: At the time of submission of the manuscript, no subjects have been recruited. The survey study was approved by the institutional ethics committee and funded in May 2025, and we started recruiting respondents in May 2025. We plan to complete phase 1 by September 30, 2025, and phase 2 by November 30, 2025. Anonymized survey results and their analyses are expected to be published in a peer-reviewed journal in fall 2026. CONCLUSIONS: We anticipate that data and analytical insights generated from this study will assist policy makers and AI researchers in prioritizing a data-informed sequence of developing and promoting AI prescription tools in successive regions, disciplines, and clinical use cases and inform policy makers to match resource allocation with "AI readiness." INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/76009.
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 | 0.040 | 0.025 |
| Meta-epidemiology (narrow) | 0.003 | 0.003 |
| Meta-epidemiology (broad) | 0.004 | 0.005 |
| Bibliometrics | 0.005 | 0.005 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.024 | 0.004 |
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".