Firstly, do no (net) harm. ~ (adapted from) Hippocrates Reading Hunink et al (2014) this morning…

🧐

So a recent Veritasium video on natural selection explains kin selection and does the unavoidable thing of saying that a parent shares "half of its genes with the child". This is wrong, because for any 2 humans, we share almost all of our genes. We share more than 95% with chimpanzees ffs. >>

This honestly should be required reading for anyone undertaking (or evaluating) exploratory studies... ⬇️👍 (It's proper funny too 😅)

NEW paper "Clinical prediction models using machine learning in oncology: challenges and recommendations" --> tinyurl.com/2h2mcy5h #machinelearning #prediction #statssky #MLsky #methodsmatter

Cool cake

Which reminds me of this old classic on the Noise Miners. “'Noise mining is a funny thing,' he said. 'When you first see a bit of noise, it doesn’t look so impressive. But as you work it out of the rock, it gets more and more refined.' This process, he explained, is called 'shucking.'"

Remembering the most likely mediator of reduced mortality in the original EGDT therapy was the presence of the CI in the intervention arm, the regular attendance of an interested and informed clinician (and skilled ICU nurse) likely has a bigger impact than any single drug or intervention.

💬 Editorial: The BICARICU-2 trial finds IV sodium bicarbonate may delay or reduce kidney replacement therapy in #criticallyill patients with severe #acidemia and #AKI, but its effect on mortality remains uncertain. #LIVES2025 @esicm.bsky.social ja.ma/47NwK8E

ANDROMEDA-SHOCK-2 is up next…. this looks at personalised phenotype-based, capillary refill time (CRT) targeted resuscitation in early septic shock, comparing to standard resuscitation. It is large (86 centres, 19 countries). Read it free from JAMA: tinyurl.com/mjcj8s5w #Lives2025 #criticalcare

EVERDAC is hot off press @esicm.bsky.social #Lives2025 in @nejm.org open access currently: www.nejm.org/doi/full/10.... Reports that in (non-blinded trial) patients with shock, management *without* early a-line insertion was noninferior to early catheter insertion…. #Lives2025 #criticalcare

In light of the art line trial. Interesting result. Need to read in depth. We really have gone full circle. A lot reflects on my practice and how I have grown up doing icu 2016 on wards and the UK practice. Back in those days, icu is a place where things are done “properly” @drfreeze.bsky.social 1/n

📚 Development and Validation of the Sequential Organ Failure Assessment (SOFA)-2 Score: Presented today in the #LIVES2025 Hot Topics session. 🔗 Free to read online: ja.ma/4qzouAz

From @jamanetworkopen.com: The SOFA-2 score updates the 1996 Sequential Organ Failure Assessment, reflecting current definitions, variables, and thresholds to grade organ dysfunction in critically ill adults receiving intensive care. #LIVES2025 @esicm.bsky.social ja.ma/3JuguA7

Who’s watching

We now have a stream of garbage in US newspapers, like this one in @washingtonpost.com denigrating London, all coded attacks on @zohrankmamdani.bsky.social Well, I live in London, love it, and am delighted to have Sadiq Khan @london.gov.uk as my mayor. www.washingtonpost.com/opinions/202...

Because 40 minutes and no slides is my ideal session length and format…. I’m back for another clinical debate! This time on considerations around timing of intubation for invasive ventilation for patients already on non-invasive treatment 🫁 @esicm.bsky.social #Lives2025

🌟 Congratulations to Professor Maurizio Cecconi, recipient of the 𝟮𝟬𝟮𝟱 𝗘𝗦𝗜𝗖𝗠 𝗦𝗼𝗰𝗶𝗲𝘁𝘆 𝗠𝗲𝗱𝗮𝗹 for his outstanding contributions in #IntensiveCare! As ESICM President (2020–2022), he led with vision through COVID-19 & drove initiatives like C19_SPACE, ICF, Airspace & the first Datathons. #LIVES2025

Spot on

Great advice as always

Through the looking glass! This blog talk about having to pay for “AI humaniser” because your original written text comes across as “AI generated” : ai.gopubby.com/when-ai-dete...

[dies of old]

😱

Well done Shafaq Aftab. When a reviewer stole her work and published it under his own name, she fought for a YEAR to get the journal to retract the paper. Just wondering why it took Wiley a YEAR to do this. 🧪 #academicsky retractionwatch.com/2025/10/03/w...

if I was a confounder I would simply be easy to observe

Here are some thoughts about higher education - specifically statistics & research methods - after grading some more papers written with "assistance" from LLMs. 🧵

Applications are open now for our new funded leadership programme Future leaders in anaesthesia Participants will gain practical, people-centred tools, expand their networks, and leave ready to take the next bold step in their leadership journey buff.ly/lF6rBLN

Aide memoirs - empty list =\= none. Forgetting to return. Randomly breaking and randomly working debug

UKRIO @ukrio.bsky.social is introducing an Authorship Integrity Toolkit: a set of practical, adaptable resources designed to support both individual researcher contributors in their day-to-day practice and organisations in developing policies and managing projects. ukrio.org/resources/th...

Mother: I have twins. Doctor: Oh, so they were born on the same day? Mother: No, they were born with the same propensity score.

@totemecho.bsky.social give these guys a fellow. Doing a great job to liberalise echo education!

Sepsis survivors identified by @stephptaylorclt.bsky.social may be useful in ARDS — and a deep dive into making the proliferation of latent subtype models clinically useful I am very proud of Dr Flick’s work

By using large language models as R data.table coding assistants, there are even fewer reasons to avoid the elegance and speed of data.table for data manipulation/aggregation/complex longitudinal data processing: hbiostat.org/rflow/long#s... #rstats

Lovely study featuring giants of ICM like @iwashyna.bsky.social Intensivist ability to prognosticate who will develop persistent critical illness is somewhat… lacking www.atsjournals.org/doi/10.1513/...

Me: oh god, I'm just an imposter! I don't actually know any statistics! <I post some mildly-hot take and it gets a like from a real statistician> Me: okay, I guess I know this one thing? Maybe?

If you're a surgeon and your anaesthetic colleagues annoy you, just walk into a governance meeting and ask them what the general opinion on cricioid pressure is, then leave. They won't notice you've gone, I promise.

The non inferior can of worms

👏🏽👏🏽👏🏽

Fascinating. For anyone who’s used LLms not just data science peeps. xayan.nu/posts/ex-mac...

There has been a lot of debate recently about the promise of real world data - the routine (observational) data collected on patients eg  treatments received, clinical outcomes etc – for estimating treatment effects. But can they deliver? 1/9 #MethodologyMonday #123

😍

TLDR: big massive dose of PPI does not work in reducing daily SOFA score or mortality. Hats off to trial-ists as it is a lot of work to set up and run properly; journals.lww.com/ccmjournal/f...

As I come to the end of my training, and my 27th and final rotation, I have reflected on this rotational aspect. It has made me extremely adaptable and provided a rich background of experience. But equally, it has broken me, prevented me from creating the social connections that are so... human.

ICU Snapshots: You start your week of dayshifts and upon entering an intubated patient's room for the first time at 7:45 am, you see this 👇 on the ventilator screen:

Before reading any article about microbiome biomarkers, see this: Zhang, J.,et al (2025). Identifying reproducible biomarkers for microbiome association studies requires thousands of sample sizes. Science Bulletin, 70(13), 2056–2060. doi.org/10.1016/j.sc...

100% this, null results are valuable and I don’t trust the people upthread saying that they aren’t!

🎓Sign up for the #LIVES2025 𝗘𝗖𝗟𝗦/𝗘𝗖𝗠𝗢 workshop, jointly organised with EuroELSO & boost your V-A and V-V ECMO skills, circuit components & troubleshoot! 📣 Enrolled in a pre-congress course? Enjoy a discount on any workshop! 𝗘𝗮𝗿𝗹𝘆 𝗳𝗲𝗲 𝗿𝗲𝗴𝗶𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻 𝘂𝗻𝘁𝗶𝗹 𝗔𝘂𝗴 𝟮𝟴!👉 www.esicm.org/events/38th-...

This article provides an overview of the current state of handling continuous variables in healthcare research. It discusses the potential limitations of assuming a linear relationship between independent and dependent variables www.bmj.com/content/390/...