META-ANALYSIS OF PATIENT-LEVEL CLINICAL TRIALS DATA
Bibliographic record
Abstract
Purpose: Patient-level meta-analysis combines data from individual patients, who participated in clinical trials, providing greater statistical power and precision compared to traditional meta-analysis. Knowledge of the patient treatment assignment could leave the analyst vulnerable to charges of bias. We provide an example of how to limit bias when conducting a patient-level meta-analysis. Methods: We developed a protocol to explore the efficacy of olsalazine for the induction of remission in patients with ulcerative colitis. We contracted the University of North Carolina to re-code the data for each randomized placebo-controlled trial with a blinded treatment indicator A or B. Since the primary outcome variable was not consistent across the trials, apanel of three experts (Brian Feagan, Canada; Derek Jewell, UK; David Sachar, US) reviewed the variables that were common to all trials and recommended a new primary outcome variable defined as absence of rectal bleeding combined with endoscopic healing at the three or four week visit as well as additional secondary outcome variables. To ensure blinding, treatment arms were balanced by dropping patients at random as follows: 1) for trials with more than two active treatment arms, only data from the arm with the highest dose were included; and 2) for trials with arms of unequal size, patient data were randomly deleted from the larger arm until both arms were of equal size. The data set was then sent to the independent analyst (LRS), who had outlined a pre-specified and approved analysis using a generalized linear mixed-effects model with random treatment effects. The final analysis was then run on the new dataset and the results were circulated to the expert panel before the blind was broken. Results: The analysis showed that olsalazine is effective for the induction of remission for ulcerative colitis (OR = 2.2; 95% CI, 1.1 – 4.4). It is also superior to placebo for eliminating rectal bleeding (OR = 2.1; 95% CI, [1.4–3.2] ) and forproviding clinical improvement (OR = 2.9; 95% CI, [1.6 – 5.1 ] ). Conclusions: This work was funded by Celltech.
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 | 0.130 | 0.283 |
| Meta-epidemiology (narrow) | 0.004 | 0.002 |
| Meta-epidemiology (broad) | 0.021 | 0.047 |
| Bibliometrics | 0.014 | 0.012 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.007 | 0.003 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.005 | 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; 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".