Uncertainty of Statistical Heterogeneity in Meta-analyses
Bibliographic record
Abstract
To the Editor: Aune and colleagues1 recently published a systematic review and meta-analysis of nonrandomized studies on the topic of physical activity and the risk of developing preeclampsia during pregnancy. The authors presented separate meta-analyses according to several measures (ie, high versus low activity, MET hours per week, and per hour of activity per day) and timing (ie, prepregnancy and during early pregnancy) of physical activity, to reduce clinical heterogeneity. Results were also stratified by study type (ie, cohort or case-control study) to address design heterogeneity. Statistical heterogeneity was assessed using Cochran’s Q, in which the P value was reported (but not the corresponding value of the χ2 statistic), and the I2 statistic, in which the value was reported (but not a 95% confidence interval [CI]). In their meta-analyses, the authors overwhelmingly found I2 values of 0% and P values exceeding 0.05, frequently interpreting these results as providing “no evidence” or “no indication” of statistical heterogeneity. However, it has been noted that such tests are underpowered,2,3 particularly when the number of studies in a given analysis is small,2 as was the case in several of the meta-analyses in this article.1 Reporting the degree of uncertainty associated with the I2 value has been recommended for clarifying any inference concerning statistical heterogeneity in meta-analyses.3 The 95% CI for I2 can easily be computed using the “heterogi” module in Stata, ideally as a reporting requirement for future authors; however, at the least, providing the value of Cochran’s Q along with the number of studies would permit calculation by interested readers. In this article, neither was provided. Deshayne B. Fell Department of Epidemiology, Biostatistics and Occupational Health McGill University Montreal Quebec, Canada Better Outcomes Registry & Network (BORN) Ontario Ottawa, Ontario, Canada Robert W. Platt Department of Epidemiology, Biostatistics and Occupational Health McGill University Montreal Quebec, Canada [email protected]
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 | 0.212 | 0.676 |
| Meta-epidemiology (narrow) | 0.003 | 0.003 |
| Meta-epidemiology (broad) | 0.011 | 0.009 |
| Bibliometrics | 0.008 | 0.008 |
| Science and technology studies | 0.001 | 0.004 |
| Scholarly communication | 0.009 | 0.007 |
| Open science | 0.008 | 0.005 |
| Research integrity | 0.015 | 0.017 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
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".