Inadequate Reporting of Harm From Randomized Clinical Trials in Top Medical Publications
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
OBJECTIVE: To assess the quality of harm reporting in randomized controlled trials (RCTs) published in high-impact general medical journals. STUDY DESIGN AND SETTING: Publications of RCTs involving drugs compared with placebo controls, that were published in five general medical journals with high Impact Factors were identified from January 2022 to December 2023. Data relating to the presentation and discussion of harm were extracted and analyzed based on the Consort Harm framework. RESULTS: We identified 175 eligible RCTs (AIM: n = 5; BMJ: n = 8; JAMA: n = 26, Lancet: n = 64, and NEJM: n = 72). None of the studies referenced the CONSORT Harms 2004 statement. Seventy-one percent of studies (n = 125) did not mention how harm data about patients' symptoms were collected and 86.3% of the analyses (n = 151) were limited to descriptive statistics. Only 45.1% of studies (n = 79) discussed the balance of benefits and harms. Common limitations included unclear methodological details, selective reporting, and inadequate analysis of results. CONCLUSIONS: RCTs published in five highly cited general medical journals contain deficiencies in harm reporting. The recently updated Consort Harm 2022 provides an implementable evaluation and guidance tool and should be actively promoted among researchers, reviewers, and journal editors. More attention to adequate and reasonable reporting requirements for harms in RCTs is necessary to provide a better opportunity for evidence-based decision making.
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.
Direct model labels (unvalidated)
Per-model category and study-design labels from the labeling rounds. They are machine output, unvalidated, and the disagreement between models ships as data. No study design here is MEDLINE-validated yet.
| Model arm | Categories | Study design | Confidence |
|---|---|---|---|
| gemma | Metaresearch Domain: Reporting · Genre: Empirical About the Canadian research system: no · About a Canadian topic: no | Observational | low |
| gpt | Metaresearch Domain: Reporting · Genre: Review About the Canadian research system: no · About a Canadian topic: no | Systematic review | high |
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.934 | 0.993 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.091 | 0.021 |
| Bibliometrics | 0.004 | 0.006 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.005 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.019 | 0.000 |
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 it