Comments on systematic review and meta-analysis of the epidemiology of man-made vitreous fibers and respiratory health outcomes
Bibliographic record
Abstract
I read with great interest the recent meta-analysis published by McElvenny et al., who stated their objective as confirming the findings of our “previous systematic review and meta-analysis of the epidemiology of man-made vitreous fibers (MMVF) in relation to malignant disease” (Egnot et al., 2020). I appreciate the authors’ engagement with our work and welcome continued research on this important topic. McElvenny et al. performed a meta-analysis of cohort studies evaluating the association between MMVF exposure and lung cancer that yielded an elevated effect estimate (pooled relative risk [RR]=1.15; 95% confidence interval [CI] = 1.01, 1.32). The authors noted that this result was higher than the result of a similar meta-analysis of cohort studies reported by our study (pooled RR = 0.97; 95% CI = 0.86, 1.09) and attributed this discrepancy to various methodological differences. Our systematic review and meta-analysis aimed to evaluate the risk of respiratory tract cancer (i.e. larynx, trachea, bronchus, and lung) among individuals with occupational exposure to only the most common types of MMVFs (i.e. glass wool and rock/slag wool). It appears that the McElvenny et al. study included additional types of specialized MMVFs, such as continuous glass filament, and specifically focused on evaluating lung cancer risk rather than broader respiratory tract cancer risk. These important differences in study inclusion criteria and the primary outcome of interest led to different results from the 2 meta-analyses. In addition to our primary meta-analysis of respiratory tract cancer risk, we also reported findings from a meta-analysis of studies that met our inclusion criteria and estimated lung cancer risk. This more appropriate comparison of findings from the McElvenny et al. and Suder Egnot et al. meta-analyses of lung cancer risk reveals that the 2 independent studies yielded quite similar results (pooled RR = 1.15; 95% CI = 1.01, 1.32 versus pooled RR = 1.10; 95% CI = 0.93, 1.30, respectively). Based on the methods provided by McElvenny et al., we suspect that this slight difference in our respective results is likely attributable to differences in our prioritization of effect estimates for inclusion in the meta-analysis (e.g. we prioritized effect estimates excluding participants with prior known asbestos exposure, and McElvenny et al. prioritized effect estimates statistically adjusted for asbestos exposure, where possible). I respectfully submit this clarification in the interest of supporting accurate interpretation of the evidence and agree with McElvenny et al. that new evidence from the ongoing large cohort studies would offer critical insights into the potential long-term human health impacts of occupational MMVF exposure. The North American Insulation Manufacturers Association has sponsored an ongoing research contract with Stantec. The funding association had no involvement or influence in the writing or conclusions of this Letter. N.S.E. is employed by Stantec, a consulting firm that provides scientific advice to government, corporations, law firms, and various scientific/professional organizations. No data were used in this study.
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 | 0.172 | 0.532 |
| Meta-epidemiology (narrow) | 0.003 | 0.004 |
| Meta-epidemiology (broad) | 0.007 | 0.017 |
| Bibliometrics | 0.007 | 0.006 |
| Science and technology studies | 0.003 | 0.008 |
| Scholarly communication | 0.006 | 0.008 |
| Open science | 0.012 | 0.005 |
| Research integrity | 0.037 | 0.032 |
| Insufficient payload (model declined to judge) | 0.008 | 0.004 |
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".