Efficacy of Risk Prediction Models and Thresholds to Select Patients for Lung Cancer Screening
Bibliographic record
Abstract
PurposeScreening for lung cancer is recommended to reduce lung cancer mortality, but there is no consensus on patient selection for screening in Canada. Risk prediction models are more efficacious than the screening recommendations of the Canadian Task Force on Preventive Health Care (CTFPHC), but it remains to be determined which model and threshold are optimal. MethodsWe retrospectively applied the PLCO m2012 , PLCO all2014 and LLPv2 risk prediction models to 120 lung cancer patients from a Canadian province, at risk thresholds of ≥ 1.51% and ≥ 2.00%, to determine screening eligibility at time of diagnosis. OncoSim modelling was used to compare these risk thresholds. ResultsSensitivities of the risk prediction models at a threshold of ≥ 1.51% were similar with 93 (77.5%), 96 (80.0%), and 97 (80.8%) patients selected for screening, respectively. The PLCO m2012 and PLCO all2014 models selected significantly more patients for screening at a ≥ 1.51% threshold. The OncoSim simulation model estimated that the ≥ 1.51% threshold would detect 4 more cancers per 100 000 people than the ≥ 2.00% threshold. All risk prediction models, at both thresholds, achieved greater sensitivity than CTFPHC recommendations, which selected 56 (46.7%) patients for screening. ConclusionCommonly considered lung cancer screening risk thresholds (≥1.51% and ≥2.00%) are more sensitive than the CTFPHC 30-pack–years criterion to detect lung cancer. A lower risk threshold would achieve a larger population impact of lung cancer screening but would require more resources. Patients with limited or no smoking history, young patients, and patients with no history of COPD may be missed regardless of the model chosen.
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 | 0.042 | 0.122 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".