Impact of aggregating exposure information from cases and controls when building a population-based job-exposure matrix from past expert evaluations
Bibliographic record
Abstract
OBJECTIVES: To assess whether the inclusion of data from cases would bias a job-exposure matrix (JEM), we evaluated whether exposures were systematically different between cases and controls from a large historical case-control study. METHODS: Data included 10 381 jobs assessed for occupational exposure to 294 agents within a lung cancer case-control study. For each sex, 1 JEM was developed from case jobs, and 1 from control jobs: with occupation (four-digit occupational codes), time period (1945-1959, 1960-1984, 1985-1995) and agent axes. We estimated concordance in exposure status (defined as probability of exposure threshold ≥5%) and exposure metrics of probability and intensity of exposure, between the 2 JEMs. RESULTS: Of all hypothetical occupation-period-agent combinations, most had no or few observations. Among males there were 8136 common cells (24-occupational codes, 3-periods, 226-agents), containing sufficient observations for comparison with 92% concordance in exposure status; discordance was equally likely to be towards cases or controls. Females had 1710 common cells (9-occupational codes, 3-periods, 114-agents) with 93% concordance in exposure status; discordant cells were more likely to reflect greater exposure among cases. Among concordantly exposed cells, probability and intensity of exposures were highly correlated between the case JEM and control JEM (Kendall τ>0.50), and absolute differences were small (median difference in probability <1.5%, median ratio in intensity=1.00) for both sexes. CONCLUSIONS: Agreement between the case JEM and control JEM was high, suggesting that aggregating the case and control information in our study into a single JEM is justifiable given the benefits of increased sample size.
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 | 0.236 | 0.427 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.009 | 0.006 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.005 | 0.003 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".