O-42 A novel weighting approach to addressing healthy worker survivor bias
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
<h3>Introduction</h3> Radon gas is a major source of ionizing radiation exposures in humans that contributes to the global burden of lung cancers. Human carcinogenicity of radon has been established, in part, in studies of exposed workers, including uranium miners. Impact estimates from occupational studies are subject to healthy worker survivor bias, which has been proposed to result in substantial underestimates of radon’s health effects. However, existing analytic methods for addressing bias due to healthy worker survivor bias are sensitive to model misspecification. <h3>Material and Methods</h3> We describe a new approach for estimating health effects of occupational exposures that addresses healthy worker survivor bias while reducing modeling assumptions. This approach utilizes inverse probability weighting and originates from the literature on dynamic treatment regimes. We use this approach to estimate impacts of hypothetical occupational standards on lung cancer mortality using data from 4124 miners from the Colorado Plateau Uranium Miners’ cohort followed through 2005. <h3>Results</h3> The estimated cumulative lung cancer mortality risk at age 80 was 14.9% (95% confidence interval [CI] = 13.7%, 16.1%). Under a hypothetical intervention to limit exposure to 20 working levels, we estimated a risk reduction (at age 80) of 2.7% (95%CI = 3.6%, 1.7%). Estimates at lower exposure levels were larger but subject to greater uncertainty than previous analyses in this cohort using modeling-based estimators. <h3>Conclusions</h3> Our approach offers substantial strengths when addressing healthy worker survivor bias, namely regarding computational simplicity and reduced reliance on modeling assumptions. Use within this highly exposed cohort also highlighted challenges with using our approach to estimate effects at low exposure levels: model-based extrapolation with the parametric g-formula can be used to reduce uncertainty under stronger assumptions. The proposed approach provides a simple approach to addressing healthy worker survivor bias that provides promise for reducing modeling assumptions in studies of occupational exposures.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it