LightNews LightNews
gladstonebrookes.bsky.social
@gladstonebrookes.bsky.social
55 followers 110 following 87 posts
luv soy, luv seed oils, hate animal abuse. simple as
Posts Replies Media Videos
Reposted
hansonmark.bsky.social
We've got ISSUES. Literally.

We scraped >100k special issues & over 1 million articles to bring you a PISS-poor paper. We quantify just how many excess papers are published by guest editors abusing special issues to boost their CVs. How bad is it & what can we do?

arxiv.org/abs/2601.07563

A 🧵 1/n
January 13, 2026 at 8:27 AM
Reposted
kcklatt.bsky.social
For anyone interested in the new Dietary Guidelines, here's a breakdown and my thoughts:
kcklatt.substack.com/p/ambiguous-...
Ambiguous DGAs & The Rancher's Pyramid
Tame on the surface, more underneath
kcklatt.substack.com
January 8, 2026 at 3:40 PM
gladstonebrookes.bsky.social
Meta-analyses of dietary exposures must consider energy adjustment (preprint) 🧵
arxiv.org/abs/2512.07531

In observational studies, adjustment for energy intake changes the nature of the effect being estimated – from addition (increasing or decreasing intake of that food while...
January 4, 2026 at 10:46 AM
Reposted
statsepi.bsky.social
‪It has a name now 😜

Many thanks to Ken for agreeing to put his good name to my...artwork. The image is in the public domain (CC 0), but citations to the linked documents are warmly welcomed.

✅ zenodo.org/records/1808...

✅ pubmed.ncbi.nlm.nih.gov/24452418/
Figure 1: The Rothman-Dahly Evidence Pyramid (original version) An equilateral triangle with a small blue section labelled "Thoughtful, well-conducted studies of any design" at the top, with the remaining space colored red and labelled "The other shit"
December 29, 2025 at 11:19 AM
gladstonebrookes.bsky.social
Interesting finding: A rather odd meta-analysis in which not a single study meets their inclusion criteria, and nearly 100% of the data presented is incorrect.

I have genuinely no idea how the heck you could end up with this (AI hallucinations? Something else?)

pubpeer.com/publications...
A screenshot from Pubpeer of the following text, titled Summary. This is described as a meta-analysis of 9 RCTs comparing Dienogest to the Combined Oral Contraceptive Pill for treatment of dysmenorrhea, dyschezia, and dyspareunia in deep endometriosis. However, 4 of these studies are not RCTs. Only 4 are comparing Dienogest to any form of COCP. Of these, none are the form of COCP reported in the meta-analysis (and two are using a COCP that also contains Dienogest). Only 1 study is specifically on deep endometriosis - the others are instead on general endometriosis, endometrioma, or adenomyosis. None of the studies have the correct number of participants reported. None of the studies have the correct mean age reported. Of the 54 data points reported in the meta-analysis, only 1 (i.e. 2%) is verifiable as being accurately extracted from the original studies and accurately described. The rest do not appear in the included studies at all, or are measuring different things than what is claimed in the meta-analysis. Therefore, with none of the included studies meeting the authors' basic inclusion criteria, and with almost 100% of the data on study characteristics and results being incorrect, this meta-analysis is rendered nonsensical.
December 16, 2025 at 7:24 PM
Reposted
pwgtennant.bsky.social
Thrilled to share my latest paper entitled, "Estimating Discrimination in Sentencing: Distinguishing between Good and Bad Controls"

Led by @jpinasanchez.bsky.social, the paper introduces a framework for examining discrimination in criminal justice processes.

🧵 1/10

publicera.kb.se/ejels/articl...
Screenshot of title page including the following abstract: To minimise confounding bias and disentangle warranted from unwarranted disparities, researchers examining sentencing discrimination have traditionally sought to control for as many legal factors as possible. However, over the past decade, a growing number of scholars have questioned this strategy, noting that many legal factors are themselves subject to judicial discretion and that controlling for them can introduce post-treatment bias. Here, we use directed acyclic graphs (DAGs) to provide a formal and comprehensive assessment of the different types of bias that may arise from different choices of controls. In addition, we propose a new modelling framework to facilitate the selection of controls and reflect the model uncertainty created by the trade-off inherent in judicially-defined legal factors and other factors with a similar dual causal role. We apply this framework to examine race disparities in US federal courts and gender disparities in the England and Wales magistrates’ court. We find substantial model uncertainty for gender disparities and for race disparities affecting Hispanic offenders, rendering estimates of the latter inconclusive. Disparities against black offenders are more consistent and — under specific conditions — could be interpreted as evidence of direct discrimination.
December 8, 2025 at 10:20 AM
Reposted
statsepi.bsky.social
We did a thing. 😬
cellpress.bsky.social Cell Press @cellpress.bsky.social · Nov 13
The link between the gut #microbiome and autism is not backed by science, researchers say.

Read the full opinion piece in @cp-neuron.bsky.social: spkl.io/63322AbxpA

@wiringthebrain.bsky.social, @statsepi.bsky.social, & @deevybee.bsky.social
First page of Opinion piece: "Conceptual and methodological flaws undermine claims of a link between the gut microbiome and autism"
November 13, 2025 at 5:59 PM
Reposted
gidmk.bsky.social
Remember that cruciferous vegetables and cancer study I wrote about recently?

Turns out it's even more borked than I said. Some excellent sleuthing here, the study probably needs to be retracted or extensively corrected.
gladstonebrookes.bsky.social gladstonebrookes.bsky.social @gladstonebrookes.bsky.social · Sep 21
Hey @gidmk.bsky.social, following this post, I dug into the study further, and it turns out there are a *lot* more issues with it.

Only one of the estimates used is actually what they claim it is (odds ratio for the association between highest vs lowest intake of cruciferous veg and colon cancer)..
September 22, 2025 at 1:14 AM
gladstonebrookes.bsky.social
Hey @gidmk.bsky.social, following this post, I dug into the study further, and it turns out there are a *lot* more issues with it.

Only one of the estimates used is actually what they claim it is (odds ratio for the association between highest vs lowest intake of cruciferous veg and colon cancer)..
September 21, 2025 at 8:59 AM
Reposted
sophieehill.bsky.social
My thread on the Neurology paper on sweeteners & cognitive health: skywriter.blue/pages/did:pl...

And a follow-up thread on some weird patterns in the Appendix tables: skywriter.blue/pages/did:pl...

Collated by Skywriter onto a web page for easier reading!
September 5, 2025 at 9:56 PM
Reposted
kcklatt.bsky.social
This year's Annual Review of Nutrition is out and full of really great reads.

My contribution details the broad history of nutritional guidance in America & covers the forces (science/politics/culture) that have driven the evolution of nutrition guidance.

www.annualreviews.org/content/jour...
August 25, 2025 at 3:49 AM
Reposted
erikarnesen.bsky.social
Nice summary of our new paper in the American Journal of Clinical Nutrition on milk consumption and cardiovascular disease mortality in #Norway 🥛
medicalxpress.com/news/2025-08... #nutrition #cardiology
Semi-skimmed milk introduction in 1980s transformed public health outcomes, study shows
A new study reveals that individuals who enjoyed whole milk during the 1970s and early 1980s had a higher risk of mortality. However, beginning in the mid-1980s, a significant change occurred.
medicalxpress.com
August 24, 2025 at 7:05 PM
Reposted
statsepi.bsky.social
Academic gut-health microbiome research has not really led to efficacious treatments for specific ailments that can be demonstrated in rigorous studies. But they *have* convinced the public that gut health is somehow, vaguely at the root of all illness, and food companies have profited off this.
August 14, 2025 at 8:03 AM
Reposted
medcrisis.bsky.social
I made a little video about the chap who trusted chatGPT to give him safe tips for better health. It did not go well.

Live on @nebula.tv now! If you sign up via this link, it helps me too :) If you're already a member, chatGPT said watching this will lower your cholesterol nebula.tv/videos/medli...
Medlife Crisis — Using chatGPT to Poison Yourself
We are on the precipice of vibe medicine. People using AI to give them health advice...who knows what the future holds as access to healthcare becomes more scarce?
nebula.tv
August 13, 2025 at 1:08 PM
Reposted
scientificdiscovery.dev
New article by me!

Cardiovascular disease mortality rates have declined by around three-quarters since 1950, but we rarely hear about it.

I explore some of the reasons behind the decline.
ourworldindata.org/cardiovascul...
This image depicts a line graph showing cardiovascular mortality rates in the United States from 1933 to 2023, alongside key advancements in medicine, surgery, and public health. The y-axis represents age-standardized death rates from cardiovascular disease, ranging from 0 to 600. The x-axis represents years from 1933 to 2023. The graph starts at around 600 deaths per 100,000 people in 1933 and trends downward sharply over the decades, indicating a significant decline in mortality rates. Key advancements are marked along the timeline, including the introduction of the first heart-lung machine in 1953, the first cardiac CT scan in 1977, and the first 3D-printed heart models in 2012. Footnote information states that data begins in 1933 when all U.S. states started reporting cardiovascular mortality rates, sourced from the National Center for Heart Statistics in 2020 and the CDC Wonder in 2025. The chart is published by Saloni Dattani at Our World in Data.
August 4, 2025 at 12:52 PM
Reposted
gilcarvalhomdphd.bsky.social
Saturated fat LIES (and how to see through them)

A recent study suggests saturated fat is harmless and has been unfairly demonized

Here´s what the influencers will never tell you, and how to protect your health and your loved ones
https://youtu.be/taaEOY5E5Iw
August 4, 2025 at 3:05 PM
Reposted
opalescentopal.bsky.social
With Tom Lehrer's passing, I suppose this is a moment to share the story of the prank he played on the National Security Agency, and how it went undiscovered for nearly 60 years.
July 27, 2025 at 9:01 PM
Reposted
elisabethbik.bsky.social
454 Hints That a Chatbot Wrote Part of a Biomedical Researcher’s Paper
Scientists show that the frequency of a set of words seems to have increased in published study abstracts since ChatGPT was released into the world.
#giftlink

www.nytimes.com/2025/07/02/h...
454 Hints That a Chatbot Wrote Part of a Biomedical Researcher’s Paper
www.nytimes.com
July 3, 2025 at 11:23 PM
Reposted
volts.wtf
Elon Musk casually, and illegally, dismantled USAID. According to a new study in the Lancet, that action will lead to *14 million* premature deaths between 2025 and 2030.

Again: Elon Musk is directly responsible for 14 million coming deaths. 4.5 million will be children.
Evaluating the impact of two decades of USAID interventions and projecting the effects of defunding on mortality up to 2030: a retrospective impact evaluation and forecasting analysis
USAID funding has significantly contributed to the reduction in adult and child mortality across low-income and middle-income countries over the past two decades. Our estimates show that, unless the a...
www.thelancet.com
July 2, 2025 at 8:00 PM
Reposted
jonhuang.bsky.social
My student sent me this and I can't unsee it, so I am blessing the bluesky timeline with this so that I will not be alone
A (likely) AI generated image of what appears to be a community health worker with a notebook talking with a woman in traditional South Asian clothing standing in front of a village with thatched roof houses. Faded in the background is a frequency table from ostensibly two groups. Overlaid on this image is the following text: "It wasn't a stethoscope. It wasn't a vaccine. It was a p-value... that changed everything. PublicHealth | DataDrivenImpact" The context: This comes from a linkedin post of someone telling a story that a p-value < 0.05 from a chi-square test was what was needed to motivate expansion of a public health program. Regardless of whether this story is true, it perpetuates an absolutely dangerous and incorrect belief that significance tests can or should be the key factor in driving new interventions and policies.
June 27, 2025 at 7:47 PM
Reposted
christofferbn.bsky.social
We just published a critical review of the KETO-CTA study — the first study to examine coronary plaque progression in Lean Mass Hyper-Responders (LMHRs) on a ketogenic diet.

Here’s why we believe the study’s results have been seriously misrepresented.

1/14
June 18, 2025 at 5:18 AM
Reposted
medcrisis.bsky.social
When two surgeons marry each other (troubling, but it happens), how do you figure out which one is the better operator? These things are important, we must know!! Thank goodness we have numbers to look at, because numbers never lie. Right?

🚨NEW VIDEO🚨

youtu.be/F6Y_bpE3r5c?...
How Do You Tell How Good Your Doctor Is?
YouTube video by Medlife Crisis
youtu.be
June 10, 2025 at 11:28 PM
Reposted
hadley.nz
Happy 18th birthday ggplot2! #rstats
Balloon wall art inspired by the ggplot2 hex sticker. The display features a hexagonal frame made of black balloons, filled with white balloons inside and surrounded by a background of light and dark blue balloons. A line graph with black balloon segments and round coloured nodes runs across the centre, and the word “ggplot2” is spelled out in black balloon letters below the chart.
June 10, 2025 at 11:56 PM
Reposted
epiellie.bsky.social
Having read the full MAHA report, what stands out the most to me, as an epidemiologist, is how inexplicably sloppy it is.

The US government collects data on every health condition you can think of, and the report doesn’t use any of it.
June 7, 2025 at 4:19 PM
Reposted
doremus-schafer.bsky.social
Oh, it's already happened: A large Norwegian city wrote a report on school closures which cited a slew of AI-hallucinated papers - and this op-ed praised the AI for "pointing to spaces of knowledge", as a way to "inspire and invite the [future] papers we need". www.nordnorskdebatt.no/tromso-feile...
Why AI hallucinates – and why that's not the point When a language model like ChatGPT can’t find concrete documentation, it tries to fill in the blanks with what it thinks fits. It “hallucinates” – but not randomly. It builds on patterns it has learned from large amounts of text, and tries to create plausible connections. This means that it sometimes invents a source, but not necessarily a bad idea. Perhaps this is precisely what we should listen to. When AI shows us connections we didn’t know were written down, it is effectively saying: “Based on everything I’ve seen, someone should have said this.” It points to a space for knowledge, not necessarily a mistake. Several researchers were upset at being attributed opinions they hadn’t formulated. That’s understandable. But it’s also possible to see it as inspiration – as invitations to precisely the reports we actually need.
May 30, 2025 at 5:44 PM