An approach to quantifying the potential importance of residual confounding in systematic reviews of observational studies: A GRADE concept paper
Bibliographic record
Abstract
Small relative effect sizes are common in observational studies of exposure in environmental and public health. However, such effects can still have considerable policy importance when the baseline rate of the health outcome is high, and many persons are exposed. Assessing the certainty of the evidence based on these effect sizes is challenging because they can be prone to residual confounding due to the non-randomized nature of the evidence. When applying GRADE, a precise relative risk >2.0 increases the certainty in an existing effect because residual confounding is unlikely to explain the association. GRADE also suggests rating up when opposing plausible residual confounding exists for other effect sizes. In this concept paper, we propose using the E-value, defined as the smallest effect size of a confounder that still can reduce an observed RR to the null value, and a reference confounder to assess the likelihood of residual confounding. We propose a 4-step approach. 1. Assess the association of interest for relevant exposure levels. 2. Calculate the E-value for this observed association. 3. Choose a reference confounder with sufficient strength and information and assess its effect on the observed association using the E-value. 4. Assess how likely it is that residual confounding will still bias the observed RR. We present three case studies and discuss the feasibility of the approach.
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 | 0.487 | 0.698 |
| Meta-epidemiology (narrow) | 0.005 | 0.005 |
| Meta-epidemiology (broad) | 0.011 | 0.020 |
| Bibliometrics | 0.044 | 0.021 |
| Science and technology studies | 0.004 | 0.011 |
| Scholarly communication | 0.016 | 0.016 |
| Open science | 0.011 | 0.015 |
| Research integrity | 0.012 | 0.013 |
| Insufficient payload (model declined to judge) | 0.006 | 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; 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".