Notifications
Clear all
Reading the evidence
2
Posts
1
Users
0
Reactions
7
Views
Topic starter
22/03/2021 3:36 am
Most arguments on peptide boards are really arguments about what a single paper showed. Before posting a study, run it through these questions and put the answers in your post.
- Who or what was studied? Cells, rodents, healthy volunteers or patients. A result in one does not automatically transfer to another.
- What was compared? Placebo, an active comparator, or nothing. Uncontrolled before/after designs cannot separate an effect from time, regression to the mean or expectation.
- What was the pre-specified primary outcome? Check the trial registration (ClinicalTrials.gov, EU CTR, ISRCTN) and compare it with what the paper headlines.
- How big, how long, how many dropped out? Small, short studies with heavy attrition give fragile estimates.
- Effect size and uncertainty, not just p-values. Quote the estimate with its confidence interval.
- Was it reported to a standard? CONSORT (trials) and ARRIVE (animal studies) list what a complete report contains. Missing items are worth noting.
- Has anyone replicated it, corrected it or retracted it? Search the DOI in the Retraction Watch database and check the journal page for notices.
- Who funded it, and who sells the product? Not a reason to dismiss a study, but a reason to read the methods closely.
Ioannidis's essay on why many published findings are false is still the shortest explanation of why small studies, flexible analyses and hot fields need extra caution.
Reply with a paper you have run through the checklist, or with a question that should be on it.
Sources
- Hopewell S, Chan A-W, Collins GS, et al. CONSORT 2025 statement: updated guideline for reporting randomised trials. BMJ 2025;389:e081123. doi:10.1136/bmj-2024-081123
- Percie du Sert N, Hurst V, Ahluwalia A, et al. The ARRIVE guidelines 2.0: updated guidelines for reporting animal research. PLoS Biol 2020;18(7):e3000410. doi:10.1371/journal.pbio.3000410
- Ioannidis JPA. Why most published research findings are false. PLoS Med 2005;2(8):e124. doi:10.1371/journal.pmed.0020124
- Retraction Watch database, openly distributed by Crossref since 2023. retractiondatabase.org
Topic starter
22/03/2021 4:48 am
Editorial follow-up
For a worked example, bring a DOI and quote the population, comparator and primary outcome exactly. Add one sentence explaining a limitation. A reporting checklist helps identify missing information; it does not certify that a study is valid.
Share: