"A New Hope: SHARE-CTD trilogy concludes with data reuse" The @sharectd.bsky.social project concludes its "trilogy" with a guide to clinical trial data reuse. In this chapter, @giulia-varvar.bsky.social introduces principles to support effective data reuse. restores.univ-rennes.fr/highlights/n...

New post for ResToRes. Real-world datasets are shaping clinical claims, but data patterns and transparency issues raise serious concerns. Our JAMA Psych letter with @ioanaacristea.bsky.social highlights several red flags in a study using the TriNetX database. restores.univ-rennes.fr/highlights/a...

The 1st RestoRes external committee meeting was held in Paris (September 2025). It brought together project partners and the external committee to discuss progress, share insights, and plan the next steps of the RestoRes project. restores.univ-rennes.fr/highlights/f...

1st public event in Paris for RestoRes (9.9.25) with a playful quiz inspired by the cult TV show Burger Quizz (some questions are a tribute to @gcabanac.cpesr.fr), followed by a roundtable with experts: what does a culture of responsible research look like? restores.univ-rennes.fr/highlights/q...

The Peer Review Congress celebrates the life of Drummond Rennie. His intellect and vision created the Congress and the field of meta-research, and sustained both for over 40 year. He was a champion of inclusion, a consummate teacher, and a wonderful friend who enriched us all. Godspeed Drummond.

Drummond’s family invites those wishing to honor his memory to make a gift to the Rogue Valley Manor Foundation crm.bloomerang.co/Ho... or Earth Justice earthjustice.org/don... @jama.com @bmj.com

🎥 The new LORIER webinar featuring Ulrich Mansmann from SHARE-CTD is now online. "Training the Next Generation of Researchers in Clinical Data Sharing" www.youtube.com/watch?v=6tx1...

Great blog by @ndevito1.bsky.social on his UKRI metascience project, which I’m honoured to join, helping authors share their code to minimum best practice standards with @bmj.com. When admin is sorted, I will hire a postdoc for the French part of the project... Keep in touch...

This might be worth inspecting further @ianhussey.mmmdata.io @floriannaudet.bsky.social @ioanaacristea.bsky.social

That’s a wrap on the 10th Peer Review Congress! Thanks to all for attending & following - we will repost in the following days & weeks to keep it all alive. Thanks to contributors and superfans @lexbouter.bsky.social @elisabethbik.bsky.social @abalkina.bsky.social and others for the amplification!

Among 134 higher-impact journals, 36 required, 98 recommended #datasharing. Most studies intended to share data for purposes and by mechanisms at the discretion of and subject to approval by the PI, without supporting documents or specified timeframes

@asapresearch.parkinsonsroadmap.org’s #OpenScience policy and compliance workflow — requiring grantees to post preprints, deposit research outputs, unambiguously i.d. research inputs — led to increases in all outcomes and activities. peerreviewcongress.o...

Journals with a purely #OA publishing model and higher impact factor were more likely than others to adopt RCT transparency policies (trial reg, protocol submission, #DataSharing) @utoronto.ca @wchospital.bsky.social @tmu-medicine.bsky.social #OpenScience #Metaresearch

Full abstract at peerreviewcongress.o...

Full report published at jamanetwork.com/journals/jam...

Full abstract at

Nonregistration, premature discontinuation due to poor recruitment, and nonpublication of #RCT results remain major challenges, especially for nonindustry trials @schwenkej.bsky.social @clinepi-basel.bsky.social @unibas.ch @mcmasteruniversity.bsky.social #OpenScience #MetaResearch #PeerReview

Full abstract available at

Comparing improperly vs properly registered RCTs submitted to @bmj.com, 2019-23, proper registration was associated w larger sample size, more authors, nonprofit funding. Trials with authors from Asia had higher odds of improper registration, per @eloder.bsky.social and #BMJ colleagues

In 2014-23 China experienced rapid growth in RCTs, surpassing Japan & the US. Neoplastic, CV, metabolic disease accounted for the largest percentages, each representing ~15-25% of all registered trials. In the US, mental health trials increased, reflecting growing interest in the area.

A @bmj.com RCT is comparing effects of use of an #OpenScience checklist on provision of study protocols, code, data, more. Too early to assess reproducibility FX! @mariskaleeflang.bsky.social @ingestegeman.bsky.social @floriannaudet.bsky.social @mlrethlefsen.bsky.social @amsterdamumc.bsky.social

SHapley Additive exPlanations (#SHAP) is a method to explain output from black box #MachineLearning models. SHAP identified words in the #McMASTERPLUS #RCT database associated w #LLM assessments of RCT rigor, quality (see image) @mcmasteruniversity.bsky.social @muala.bsky.social

Full abstract available at

#ChatGPT was accurate in identifying changes in RCT outcomes on ClinicalTrials.gov, and completed the entire process (ie, matching, defining, comparing) in ~2 min per trial vs 27 min for humans @ischoolui.bsky.social @colbyvorland.bsky.socia #clinepi #metaresearch peerreviewcongress.o...

Risks, barriers to #OpenResearch in UK higher ed id’d in the 2022-23 UK #Reproducibility Network Open & Transparent Research Practices Survey were personal (eg error), systems/institutional (eg resources), ethical/quality @lhughesnoehrer.bsky.social @naubertbonn.bsky.social @manchester.ac.uk

#GPT-4o performed admirably at assessing whether published #RCTs include the information recommended by #SPIRIT 2013 and #CONSORT 2010 reporting guidelines @emilyricotta.bsky.social @deloittedigi.bsky.social

#ZakKohane on #AI in #PeerReview: the surge in scientific publication is straining the conventional review system beyond capacity. @ai.nejm.org piloting AI-augmented fast track review: manuscripts w high prior of being good gets assigned to #GPT5 + a human, disposition reviewed by editorial team!

#NLP comparison of #COVID19 early-#pandemic pre- vs post-prints identified 22.6% (n=105) that had changes in final form that could potentially change policy guidance, including conclusions about the effectiveness of clinical or nonpharma interventions @emilyricotta.bsky.social

Full PRC abstract at

In a non-randomized comparison of a 21- v 14-d #PeerReview return deadline involving 593 review invitations for 95 JCO #Oncology Practice manuscripts, late reviews decreased (from 27% to 22%) & median time to review submission increased (from median 19 to 13 d) @ascocancer.bsky.social #metaresearch

Full PRC abstract available at

Ethical review boards in Sweden handled 20,440 applications 2021-23, mostly amendments, primarily from health care systems (51%), universities (42%), & corporations (3%). COVID-19 studies had expedited processing times (median t to decision, 51 vs 63d; P < .001). @stockholm-uni.bsky.social @ki.se

Full abstract available at

In a @critcaremed.bsky.social quasi-RCT comparing financial incentives for #PeerReview ($250 vs $0), there was no difference in % invites accepted (48 vs 53%) but paid reviewers were more likely to complete reviews (OR 1.36; 95% CI. 1.01-1.84) journals.lww.com/ccm...#PeerReview@sccmcriticalcare.bsky.social

Full PRC abstract at

In a @bmj.com survey of 183 patient/public #PeerReviewers assaying their experience, perspectives on payment (£50 or an online #BMJ journal subscription) (69% response rate), 84% reported a good/very good experience, 48% might be more likely to review @emmajdoble.bsky.social @amyprice.bsky.social

Full PRC abstract available at

Among 159 #ChineseMedicalJournal peer reviewers who responded to an online survey about AI use (42% response rate), 93 (58.5%) reported use of AI tools in writing and preparing papers, particularly for language translation and polishing #peerreview @wolterskluwer.com

Comparing submitted vs published RCT abstracts eval'd at @nejm.org in 4 domains (design, outcome, AEs, conclusions). published abstracts improved in a mean 0.9 domains. More improvements observed in leading gen med vs 433 other journals (72 % vs 48%) www.acpjournals.org/... @cpkotanidis.bsky.social

Full PRC abstract at

Paper characteristics associated w being sent for review, final acceptance @science.org #ScienceAdvances 2015-19 include institutional prestige, team size, ms topic, country, gender. Expert advice also correlates w final editorial decision @aaronclauset.bsky.social @sfiscience.bsky.social @aaas.org

Full abstract available at

Comparing 1291 preprint vs published versions-of-record (VORs) Oct ’22 - Dec ’24 under @eLife's Publish-Review-Curate model (where manuscripts are published as publicly reviewed preprints), 82.6% were declared ‘VOR’ after 1 revision w changes to strength, significance of evidence most common.

Full PRC abstract at

Full abstract at

In the med literature, statements like “to our knowledge [TOK], this is the FIRST...” often replace a proper systematic search. Among 4488 PMC articles in 6 leading medical journals published 2020-2024, TOK prevalence was 21.7%, w @PLOSMedicine highest (48.4%) and @NEJM lowest (1.3%) #CornellVet

Full abstract at

2.2.1 In an RCT comparing 'uselessly elongated' vs shorter #PeerReview of @neuripsconf.bsky.social 2022 submissions, the longer reviews received higher quality scores, suggesting a length bias among reviewers, w implications for #LLM-generated reviews journals.plos.org/pl...#PeerReview#LLM-generatedscsatcmu.bsky.social

In an RCT comparing anonymous vs non-anonymous #peerreview discussions of #UAI22 conference abstracts, there were modestly more discussion posts and reviewers felt safer expressing their opinions w anonymous review journals.plos.org/pl...#peerreview#UAI22scsatcmu.bsky.social #AI #CompSci

The transition of The ALMA Observatory in 2021 from single-anonymous panel reviews to dual-anonymous distributed #PeerReview to manage submission volume was associated w reduced disparities by PI experience & region, consistent w reduced prestige & geographic bias @thenrao.bsky.social #astronomy