Bayesian adaptive enrichment design in multi-arm clinical trials: The BayesAET package for R users
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
BACKGROUND: Randomized controlled trials seldom assess treatment effect heterogeneity across subpopulations, potentially leading to suboptimal treatment recommendations and inefficient use of healthcare resources. Adaptive enrichment designs seek to identify patient subpopulations most likely to benefit from the treatment. This manuscript introduces BayesAET, an R package developed to support Bayesian adaptive enrichment trial designs. The package helps identify optimal treatments for pre-specified subpopulations within a broader patient population, improving the efficiency and relevant inference of clinical trials. METHODS: BayesAET integrates Bayesian multi-arm multi-stage designs with adaptive enrichment strategies. It allows for the incorporation of historical data through Bayesian priors, supports adaptive randomization and interim analyses. These features facilitate flexible but robust modifications to trial parameters based on accumulated data, including early stopping, dropping ineffective treatments, and adjusting randomization probabilities. The package supports various outcome types, including continuous, binary, and count outcomes. RESULTS: We showcase BayesAET through a case study of a trial evaluating repetitive transcranial magnetic stimulation for depression and anxiety. The trial involved three treatment protocols and two subpopulations (with and without benzodiazepine use). Simulations demonstrate that BayesAET effectively identifies differential treatment effects, adapts trial parameters based on interim data, and improves precision in treatment effect estimation. CONCLUSION: BayesAET provides a comprehensive tool for designing and analyzing Bayesian adaptive enrichment trials to identify the optimal treatments with pre-specified subpopulations.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.124 | 0.082 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it