How do network meta-analyses address intransitivity when assessing certainty of evidence: a systematic survey
Bibliographic record
Abstract
OBJECTIVES: To describe how systematic reviews with network meta-analyses (NMAs) that used the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) NMA approach addressed intransitivity when assessing certainty of evidence. DESIGN: Systematic survey. DATA SOURCES: Medline, Embase and Cochrane Database of Systematic Reviews from September 2014 to October 2022. ELIGIBILITY CRITERIA: Systematic reviews of randomised controlled trials with aggregate data NMAs that used the GRADE NMA approach for assessing certainty of evidence. DATA EXTRACTION AND SYNTHESIS: We documented how reviewers described methods for addressing intransitivity when assessing certainty of evidence, how often they rated down for intransitivity and their explanations for rating down. RESULTS: Of the 268 eligible systematic reviews, 44.8% (120/268) mentioned intransitivity when describing methods for assessing the certainty of evidence. Of these, 28.3% (34/120) considered effect modifiers and from this subset, 67.6% (23/34) specified the effect modifiers; however, no systematic review noted how they chose the effect modifiers. 15.0% (18/120) mentioned looking for differences between the direct comparisons that inform the indirect estimate. No review specified a threshold for difference in effect modifiers between the direct comparisons that would lead to rating down for intransitivity. Reviewers noted rating down indirect evidence for intransitivity in 33.1% of systematic reviews, and noted intransitivity for network estimates in 23.0% of reviews. Authors provided an explanation for rating down for intransitivity in 59.6% (31/52) of the cases in which they rated down. Of the 31 in which they provided an explanation, 74.2% (23/31) noted they detected differences in effect modifiers and 67.7% (21/31) specified in what effect modifiers they detected differences. CONCLUSIONS: A third of systematic reviews with NMAs using the GRADE approach rated down for intransitivity. Limitations in reporting of methods to address intransitivity proved considerable. Whether the problem is that reviewers neglected to address rating down for transitivity at all, or whether they did consider but not report, is not clear. At minimum systematic reviews with NMAs need to improve their reporting practices regarding intransitivity; it may well be that they need to improve their practice in transitivity assessment. How to best address intransitivity may remain unclear for many reviewers thus additional GRADE guidance providing practical instructions for addressing intransitivity may be desirable.
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 | 0.501 | 0.870 |
| Meta-epidemiology (narrow) | 0.003 | 0.005 |
| Meta-epidemiology (broad) | 0.013 | 0.023 |
| Bibliometrics | 0.028 | 0.020 |
| Science and technology studies | 0.002 | 0.005 |
| Scholarly communication | 0.014 | 0.026 |
| Open science | 0.004 | 0.008 |
| Research integrity | 0.006 | 0.005 |
| Insufficient payload (model declined to judge) | 0.005 | 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".