The TRACTOR Project: TRACking and MoniToring Occupational Risks in Agriculture Using French Insurance Health Data (MSA)
Bibliographic record
Abstract
Abstract Objectives A vast data mining project called ‘TRACking and moniToring Occupational Risks in agriculture’ (TRACTOR) was initiated in 2017 to investigate work-related health events among the entire French agricultural workforce. The goal of this work is to present the TRACTOR project, the challenges faced during its implementation, to discuss its strengths and limitations and to address its potential impact for health surveillance. Methods Three routinely collected administrative health databases from the National Health Insurance Fund for Agricultural Workers and Farmers (MSA) were made available for the TRACTOR project. Data management was required to properly clean and prepare the data before linking together all available databases. Results After removing few missing and aberrant data (4.6% values), all available databases were fully linked together. The TRACTOR project is an exhaustive database of agricultural workforce (active and retired) from 2002 to 2016, with around 10.5 million individuals including seasonal workers and farm managers. From 2012 to 2016, a total of 6 906 290 individuals were recorded. Half of these individuals were active and 46% had at least one health event (e.g. declared chronic disease, reimbursed drug prescription) during this 5-year period. Conclusions The assembled MSA databases available in the TRACTOR project are regularly updated and represent a promising and unprecedent dataset for data mining analysis dedicated to the early identification of current and emerging work-related illnesses and hypothesis generation. As a result, this project could help building a prospective integrated health surveillance system for the benefit of agricultural workers.
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 | 0.007 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.006 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 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 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".