Improving the trustworthiness of findings from nutrition evidence syntheses: assessing risk of bias and rating the certainty of evidence
Bibliographic record
Abstract
Suboptimal diet is recognized as a leading modifiable risk factor for non-communicable diseases. Non-randomized studies (NRSs) with patient relevant outcomes provide many insights into diet-disease relationships. Dietary guidelines are based predominantly on findings from systematic reviews of NRSs-mostly prospective observational studies, despite that these have been repeatedly criticized for yielding potentially less trustworthy results than randomized controlled trials (RCTs). It is assumed that these are a result of bias due to prevalent-user designs, inappropriate comparators, residual confounding, and measurement error. In this article, we aim to highlight the importance of applying risk of bias (RoB) assessments in nutritional studies to improve the credibility of evidence of systematic reviews. First, we discuss the importance and challenges of dietary RCTs and NRSs, and provide reasons for potentially less trustworthy results of dietary studies. We describe currently used tools for RoB assessment (Cochrane RoB, and ROBINS-I), describe the importance of rigorous RoB assessment in dietary studies and provide examples that further the understanding of the key issues to overcome in nutrition research. We then illustrate, by comparing the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach with current approaches used by United States Department of Agriculture Dietary Guidelines for Americans, and the World Cancer Research Fund, how to establish trust in dietary recommendations. Our overview shows that the GRADE approach provides more transparency about the single domains for grading the certainty of the evidence and the strength of recommendations. Despite not increasing the certainty of evidence itself, we expect that the rigorous application of the Cochrane RoB and the ROBINS-I tools within systematic reviews of both RCTs and NRSs and their integration within the GRADE approach will strengthen the credibility of dietary 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 | 0.807 | 0.954 |
| Meta-epidemiology (narrow) | 0.005 | 0.007 |
| Meta-epidemiology (broad) | 0.017 | 0.019 |
| Bibliometrics | 0.050 | 0.025 |
| Science and technology studies | 0.005 | 0.013 |
| Scholarly communication | 0.021 | 0.018 |
| Open science | 0.009 | 0.017 |
| Research integrity | 0.015 | 0.012 |
| Insufficient payload (model declined to judge) | 0.005 | 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".