Methodological Issues in Rating Certainty of Evidence and Interpreting Magnitude of Effect in Systematic Reviews and Practice Guidelines
Bibliographic record
Abstract
In the development of a BMJ Rapid recommendation – an international practice guideline initiative led by the MAGIC Evidence Ecosystem Foundation, and aiming to produce trustworthy, accessible and timely guidance – of plasma exchange and dosage of corticosteroids for patients with ANCA-associated vasculitis (AAV) (Chapter 2) two methodological issues arose. The first issue is related to the rating of the certainty of evidence supporting the recommendations. Reviewers experienced challenges in making an explicit statement about what it was in which they were rating their certainty (i.e., the target of the rating of certainty of evidence). Through iterative discussions and presentations at GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) Working Group meetings, the research team developed new GRADE guidance (Chapter 3 and 4) to help systematic reviewers be aware of the importance of determining the target of their rating of certainty of evidence and provided practical principles to help systematic reviewers specify this target. The second issue arose from the process of moving from evidence to decisions. To help the BMJ Rapid recommendation panel interpret the magnitude of benefit and harm associated with plasma exchange, which required understanding patient values and preferences, the research team created a panel survey for eliciting the panelists’ view regarding patient values and preferences. The research team then applied the panel survey approach in some other guidelines. Based on the experience of developing panel surveys, and through iterative discussions and consensus, the research team developed a framework for using surveys to guide guideline panels in making inferences regarding patient values and preferences (Chapter 5). Using interpretive description, the team conducted a qualitative evaluation regarding the influence of the panel surveys on the panels’ understanding of patient values and preferences, interpretation of magnitude of benefits and harms, and on panels’ decision on guideline recommendations (Chapter 6). The panel surveys proved to help guideline panels explicitly consider and incorporate patient values and preferences in making recommendations.
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 | 0.880 | 0.960 |
| Meta-epidemiology (narrow) | 0.005 | 0.009 |
| Meta-epidemiology (broad) | 0.017 | 0.020 |
| Bibliometrics | 0.032 | 0.035 |
| Science and technology studies | 0.006 | 0.023 |
| Scholarly communication | 0.029 | 0.022 |
| Open science | 0.013 | 0.017 |
| Research integrity | 0.019 | 0.020 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
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".