Navigating analytical challenges in clinical trials using the multiverse approach
Bibliographic record
Abstract
Making decisions regarding data processing and analysis are crucial steps toward extracting insights from data in clinical trials. Trial registries like clinicaltrials.gov promote transparency about these decisions and encourage making them in advance. However, clinical studies often face decisions with multiple reasonable options outside the bounds of preregistration, such as when studies conduct post hoc analyses, deviate from preregistered plans, or simply were not preregistered. Additionally, even a priori decisions often have multiple reasonable options from which to choose. Methods that maximize transparency and minimize bias in such situations are needed. This paper advocates for applying a “multiverse” approach to analyzing such data from clinical trials. The multiverse approach simultaneously selects and analyzes the various reasonable options for each decision and presents results across all analysis “universes.” We highlight common challenges and decisions when analyzing clinical trial data, review and expand upon the multiverse approach and show how it can address these challenges, and demonstrate the approach using data from a small randomized psychotherapy trial for posttraumatic stress disorder. In the example presented, results were fully consistent across the multiverse for one outcome (posttraumatic stress symptoms), partially consistent for another (relationship satisfaction), and mostly inconsistent for a third outcome (fear of intimacy). The multiverse approach is a flexible and transparent analysis option for clinical trials in the presence of uncertainty regarding data processing and analytic choices.
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 | 0.642 | 0.742 |
| Meta-epidemiology (narrow) | 0.005 | 0.005 |
| Meta-epidemiology (broad) | 0.014 | 0.016 |
| Bibliometrics | 0.020 | 0.011 |
| Science and technology studies | 0.004 | 0.012 |
| Scholarly communication | 0.017 | 0.012 |
| Open science | 0.008 | 0.015 |
| Research integrity | 0.008 | 0.018 |
| Insufficient payload (model declined to judge) | 0.006 | 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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".