There simply is no need about AI-based diagnosis of any mental disorder. Criticism of these nonsense papers should not obviate a more fundamental discussions. We do not need engineers and AI to profile mental disorders, no matter how they do it and write it up.

Perfect

New preprint retracted randomized trials attributed to super-retractors and top-cited scientists with multiple retractions: secondary analysis of the VITALITY retrospective cohort www.medrxiv.org/content/10.1... Led by PhD students, C Lyu & M Matbouriahi, w/ @floriannaudet.bsky.social, J Ioannidis

Your Sunday morning (in my case Saturday afternoon) heart-break www.newyorker.com/culture/the-...

Thanks to BMJ's Open peer-review fascinating to read that there was a considerable effort of peer-review invested into this trial before publication

@bmj.com Please look at PubPeer comments on an article you published last week. pubpeer.com/publications... I think your research integrity dept shld act swiftly on this one, given clinical significance. I'm aware of even more evidence of problems so let me know if this is not sufficient.

New blogpost on my concerns about Wellcome LEAP new $50m program on autism/microbiome links. deevybee.blogspot.com/2025/10/a-le... #autism #microbiome #biomarkers #diagnostic

Trying to find out what happened to an article. In theory, it's retracted. In practice, neither the article, nor the retraction notice are anywhere to be found. pubmed.ncbi.nlm.nih.gov/28991839/ Doi: 10.1097/MPG.0000000000001772 (leads nowhere). There is a copy of the accepted manuscript online.

I recently reported on an ICE officer who was publicly reprimanded and “relieved of his duties” after shoving a woman to the ground in an NYC immigration court. Well, he’s back. DHS did not respond to a request for comment. Original story: www.propublica.org/article/ice-... Photos: Carol Guzy

New writing, a short piece this time emilytammam.substack.com/p/this-daugh...

BREAKING: Judge William Young, a Reagan appointee, delivers the most scathing legal rebuke of the Trump era, ruling that Trump and his cabinet illegally targeted pro-Palestinian students for deportation to "strike fear" into First Amendment protesters. storage.courtlistener.com/recap/gov.us...

Must, must-read. Just drop everything, take an hour and read this, as I have. I wish these two parents would know that strangers like me who read their writings about their Martha will think long and hard about what happened, not forget and not be the same.

The ‘pro’ reasons to share the paper with the authors of the p-curve before sharing are an uphill journey in cringe. Really terrible. Had Amy Cuddy written these back in the day, white knight saving science twitter would have crucified her.

Some researchers don't discuss their future research plans for fear of being scooped. Not me. I drop bad ideas for unscrupulous people to 'steal'. - What are the neural correlates of Open Science practices? - What is the role of habits in learning a new skill through repetitive practice?

For the record, the local police had to arrive to free the girl from … the federal police. This is horrific. It’s undeniable that we will need trials for ICE officials and officers after all this. “Leominster police arrived at the scene, recovered the child and returned her to the family.”

Cool, cool, so acetaminophen is no longer ok in pregnancy, well we'll just use the *other* drugs on the safe list for pain (haha), fever...ah wait. Now that we fixed this, can we focus on how paternal age and widening age gaps contribute to risk of ASD? pubmed.ncbi.nlm.nih.gov/26055426/

Our piece in @psyche.co, based on our perspective @thelancet.com Psychiatry www.thelancet.com/journals/lan...

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...

A meta-perspective by Malcolm Macleod on the presentations at #PRC10. Are we going for low hanging fruit too much in research on peer review / publication?

Ioannidis: Do we enough evidence for your proposed actions to improve peer review? Macleod: The evidence is thin, partly because many journals are hesitant to accommodate meta-research, like RCTs. #PRC10

John Ioannidis: "We need more RCTs" I agree, so here is an urgent call to the representatives of journals at #PRC10: Let's empirically test suggested improvements to peer review like open reports, open identities, structured review, results-free review, collaborative review, etc. Get in touch!

Extreme, outlier citation concentration patterns can distort citation-based metrics, especially the #Hindex. While not implying misconduct they can serve as orange/red flags for further analysis. @iakevdaimon.bsky.social @polytechniqueparis.bsky.social @uni.lu Preprint at arxiv.org/pdf/2406.1...

We didn't randomize, and there was no allocation concealment or blinding, and we can't really be sure what intervention they got or how the outcomes were measured, but we emulated a trial by drawing a DAG.

As a reminder, our comment published here (sciencedirect.com/science/arti...) showed that (i) randomization never occurred; (ii) irregularities in baseline scores, which for the same students vary systematically in ways that are unique to either the treatment or control group; ...

‘Why, to avoid naming Boccaccio, did Chaucer invent a fictional Latin poet as his source for 𝘛𝘳𝘰𝘪𝘭𝘶𝘴? It seems that Boccaccio already had a reputation problem.’ Barbara Newman on Boccaccio’s dirty book: www.lrb.co.uk/the-paper/v4...

Effects of ketamine and esketamine on death, suicidal behaviour, and suicidal ideation in psychiatric disorders: A systematic review and meta-analysis https://www.medrxiv.org/content/10.1101/2025.08.19.25333796v1

Perhaps the best irony here, lost on the promoters, is that forensic science actually has really shoddy, mostly non-evidence based, often harmful practices. The police also is generally a beacon of fairness and justice.

Man, i could have written this, but probably less eloquently. Let’s start by saying I too thought things had gone awry when Science invited us to meet the data thugs. I would be very ok with *never* meeting any thugs.

Devastating for global health. “A federal appeals court panel cleared the way on Wednesday for the Trump administration to continue refusing to spend billions of dollars in foreign aid, finding that aid organizations that had sued to recover the money lacked the legal right to bring the challenge.“

1. Meta-research has been doing this literally for *decades*, with sophisticated methodology. John Ioannidis, @isabelleboutron.bsky.social and others, I myself have papers on this eg pubmed.ncbi.nlm.nih.gov/29183418/ psycnet.apa.org/record/2022-...

The p-curve has no place in meta-analyses. Now what do we do with all the papers that used a deeply flawed methodology whose main advantage was confirming pre-existing biases? If this weren’t literature from the ‘good’ side, the jacobins would say hunt down and retract. (This is not what I believe)

'Tell us how he died': Salah criticises UEFA tribute to 'Palestinian Pele' reut.rs/40XnM5d

Forget cancer, forget rare diseases, let’s shift the R&D focus to…measles for a treatment that will cost immensely to develop and likely not work (how many successful monoclonal antibodies for viruses?), while there is an effective and safe vaccine that ppl mistrust b/c of one retracted case series

Indeed. Respectful care also means 1. Treating mother well-being as a priority, in general and especially in the immediate postpartum. Stop normalizing extreme sleep deprivation. Treat mother sleep as a top priority.

Between classical ballet and fencing, my daughter is doing a great job of fulfilling *both* her parents’ thwarted childhood obsessions. Now what to project on the other one? (Twitter would not have disappointed at this one)

If I had a nickel for all the times I have said this myself, usually met with the blank ‘but they are cited’ (also in shit journals). I like this paper and I think it’s important but the doom hype has taken a life of its own. And ofc has its own industry.

Very much along my own thoughts. I think this is a great paper, gargantuan effort, rigorous, elegant methodology. I also love @carlzimmer.com s article. Why then we need this doomsday spin? Science will self-destroy? It’s 30k papers (btw not all clearly papermills) and a ring of 40-50 editors.

‘Stay with me, Sheila. It’s Prof Godfrey. You’ve lost a lot of blood, but the ambulance is on its way. Before it gets here, though, we need you to just come and change the toner cartridge in the photocopier for us. Sheila? Oh ffs...Has anyone got smelling salts?

computer is it normal for the state dept spokesperson to sound like my freshman philosophy professor

Does anyone know of a direct comparison between EU countries in terms of metrics used to evaluate universities and/or academic researchers? I'm especially interested in whether one can contrast countries that use N publications with those that don't. I suspect it's not so simple though

What a surprise (no). And what do you know, manual check is still the gold standard.

Noone warned me of unintended consequences of your children correctly using the names of *all* body parts combined with their naturally tendency of narrating everything they see, with no awareness of context.

Similar but subtly different British translations 1. “Anyway, I won’t keep you any longer” Translation: I’m going to walk away from you now because I want to go. 2. “Anyway, don’t let me keep you any longer” Translation: I have to stay here, so I can’t walk away, but I want YOU to walk away.

Very nice!

This, broadly, is kind of undeniable anymore.

Nothing more can add to these doctors’ words www.theguardian.com/global-devel...

🥁 This week, we submitted to BMC medicine our stage 2 registered report "Efficacy and safety of esketamine for ‘treatment-resistant depression’: registered report for a systematic review with an individual patient data meta-analysis of randomized, double-blind, placebo-controlled trials". 🧵

Let's set a threshold (X retractions, X "atypical" collaboration networks, X problematic co-authors etc) over which a scarlet letter is associated to your ORCID. If we are at author background check, might as well do it well www.dimensions.ai/products/all... www.digital-science.com/blog/2024/12...