Abstract 197: Multilevel Analysis Of Clinical Trial Data: Is It Necessary?
Bibliographic record
Abstract
Introduction: Major clinical trials involve any number of sites in order to enroll enough patients to adequately power the study. In addition, sites enrolling patients in a trial may belong to different health systems (e.g., countries). Differences among patients and practices of these systems/sites potentially introduces a component of variation that may affect the results of the statistical analysis of the data, and thus the conclusions of the study. In this study, we examined the impact of the multilevel nature of the COURAGE Trial on the analysis of patients’ quality of life data. Methods: The aim of the COURAGE Trial was to determine whether the addition of PCI to optimal medical therapy, when used as an initial management strategy, reduces the risk of death or nonfatal MI in patients with stable CAD, compared with optimal medical therapy alone. A total of 2,287 CAD patients were enrolled from 3 health care systems: Canadian (16), VA (15) and US non-VA (19) hospitals. A 3-level mixed effects model was used to analyze differences in change is Seattle Angina Questionnaire (SAQ) scores from baseline to 3 years follow-up. Health care system and hospital were treated as random effects, repeated measurements of patient SAQ scores was a fixed effect. Within-patient variability, as in any analysis, was considered a random effect. The intra-class correlation (ICC) coefficient was calculated as an index of the random effects on the analysis. Results: The results of the COURAGE quality of life study have been previously published. The multilevel analysis indicated that health care system and hospitals had little impact on the results. The ICC including health care system and hospital as random effects was 0.35 compared to 0.34 when not included in the analysis. Conclusions: In the COURAGE Trial, multiple health care systems and hospitals had virtually no effect on the statistical results. However, this may not necessarily be the case for all studies in which there is a multilevel component. Multilevel effects should always be assessed in these types of studies.
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 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 | 0.043 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".