Tobacco smoking and gastric cancer: meta-analyses of published data versus pooled analyses of individual participant data (StoP Project)
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
Tobacco smoking is one of the main risk factors for gastric cancer, but the magnitude of the association estimated by conventional systematic reviews and meta-analyses might be inaccurate, due to heterogeneous reporting of data and publication bias. We aimed to quantify the combined impact of publication-related biases, and heterogeneity in data analysis or presentation, in the summary estimates obtained from conventional meta-analyses. We compared results from individual participant data pooled-analyses, including the studies in the Stomach Cancer Pooling (StoP) Project, with conventional meta-analyses carried out using only data available in previously published reports from the same studies. From the 23 studies in the StoP Project, 20 had published reports with information on smoking and gastric cancer, but only six had specific data for gastric cardia cancer and seven had data on the daily number of cigarettes smoked. Compared to the results obtained with the StoP database, conventional meta-analyses overvalued the relation between ever smoking (summary odds ratios ranging from 7% higher for all studies to 22% higher for the risk of gastric cardia cancer) and yielded less precise summary estimates (SE ≤2.4 times higher). Additionally, funnel plot asymmetry and corresponding hypotheses tests were suggestive of publication bias. Conventional meta-analyses and individual participant data pooled-analyses reached similar conclusions on the direction of the association between smoking and gastric cancer. However, published data tended to overestimate the magnitude of the effects, possibly due to publication biases and limited the analyses by different levels of exposure or cancer subtypes.
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 | 0.006 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.004 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.002 |
| 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