One reason I oppose associational or other non-causal language in research questions is it lets people think they don't need causal inference to answer causal questions. (When the underlying question is causal, of course. Which it almost always is.)

We have created a resource in the ISBA website where you can find all recorded webinars: bayesian.org/ba-webinars/ Also, don't forget about the ISBA YouTube channel featuring curated playlists from ISBA World Meetings and specialized workshops and seminars: www.youtube.com/@isba-intern...

I can't be the only one, right? #statsmeme

We've all been there... 🫂

I don’t want kids “pushing to main”. What are they pushing? It could be drugs.

Hadn’t heard this before either. Interesting!

Well. I get exasperated about this. The problem is that many expect statistics to be easy. Some then get annoyed when it isn’t. Mathematics gets a free pass it seems. I am still learning statistics after more than 50 years.

I just want you guys to know that if your #rstats models aren't converging, it's likely because you're taking Tylenol. Tylenol can flatten likelihood surfaces due to multidimensional recontainerization of typhoid surfaces. I'm going to ask the FDA to modify the label of Tylenol to indicate this.

#statsmeme 🧵inspired by @dingdingpeng.the100.ci and @vincentab.bsky.social's awesome paper:

Whoa—my book is up for pre-order! 𝐌𝐨𝐝𝐞𝐥 𝐭𝐨 𝐌𝐞𝐚𝐧𝐢𝐧𝐠: 𝐇𝐨𝐰 𝐭𝐨 𝐈𝐧𝐭𝐞𝐫𝐩𝐫𝐞𝐭 𝐒𝐭𝐚𝐭 & 𝐌𝐋 𝐌𝐨𝐝𝐞𝐥𝐬 𝐢𝐧 #Rstats 𝐚𝐧𝐝 #PyData The book presents an ultra-simple and powerful workflow to make sense of ± any model you fit The web version will stay free forever and my proceeds go to charity. tinyurl.com/4fk56fc8

I knew that there has been incredible progress for diabetics, but I hadn’t been quite aware of the central role that insulin played in kicking off biotechnology as a field ✨

Idea for a question on the final exam: <Screenshot of a #stats dunk from Bluesky> Explain this dunk. Why is the quoted person wrong? What is the point made in this post? What would you do differently?

I'm inclined to agree. I haven't implemented an MCMC sampler but I've implemented the MICE algorithm (also a Gibbs sampler) from scratch and that was illuminating

The new {marginaleffects} release for #RStats (0.30.0) comes with two new vignettes: 1. Speed up computation with automatic differentiation (often 10x gains) marginaleffects.com/bonus/perfor... 2. Power analyses with {marginaleffects} and {DeclareDesign}. marginaleffects.com/bonus/power....

NEW PAPER! (in press in EPIDEMIOLOGY) Have you wondered: - How to specify a target trial: as an ideal trial or something else? - What biases do target trial emulations and actual RCTs share? - How does it all relate to potential outcomes? Read here! 👉 arxiv.org/abs/2405.10026 #EpiSky #CausalSky

lol my new favorite in-class example for Causality and DAGs day lecture just dropped 😆 #AcademicChatter #statistics

Our practice of anaesthesia has been empirically derived over the last two centuries. We have yet to identify the neurophysiological basis of anaesthesia because we have yet to identify the neurophysiological basis of consciousness. So what we do is fundamentally a sophisticated guess.

The only answer we have at present is provide more anaesthetic

3. In real world practice intraoperative (connected) consciousness is still common and is a concern for patients

We have over-promised by claiming we induce unconsciousness and verifying this by testing amnesia (which occurs at most lower doses of drugs)

2. Anaesthetists have created their own jargon on this topic and conflate memory with consciousness with “awareness”. This is confusing to other fields and professions.

This is a topic close to my heart. I share concern that I’m too verbose for this format! The primary issues are: 1. Who decides what is acceptable? The answer is the patient. Our crude attempt to quantify this in the public suggested that patients would prefer amnesia and lack of experience

I believe when working with big data for causal inference, you HAVE to pre-specify your analyses (esp within a project). Do it for yourself, not for reviewers. As Feynman said, “[in science] the first principle is that you must not fool yourself and you are the easiest person to fool” #stats

I do get a lot of joy when I get a difficult airway secured!!

Bit of a curve ball for the #MedSkyDebate this weekend . Nicholas Chrimes @chrimesy.com says use a MAC 3. What do you you do?? The @VortexApproach has had such an impact around the world. Do you use it? Dual debate questions. https://youtu.be/gYxwhEmYb9w

Following on the weekend #MedSkyDebate is a continuation of the Regional theme. “What do you do to improve the experience of regional anaesthesia for your patients?” Thanks again to @benmoran.bsky.social for this really interesting topic. Music? Sedation? Chatter? GA? It can be scary for patients.

Well of course this brings us to the Japanese concept of kaizen - the asymptotic approximation of perfection through small incremental gains over time.

I've occasionally been asked in job interviews "what's the best anaesthetic you've ever given?" Each time I fumbled a mediocre answer. If I get asked again, I'll say "I don't know - it hasn't happened yet. But I'm getting ready for it"

I think is an important point about research into different anaesthetic techniques. Firstly, there’s the issue of heterogeneity of treatment effect. Where does your pt sit within the observed outcome range?

I agree Dave. We need to balance what is best for each patient. I think @benmoran.bsky.social was also highlighting what to do when the block wears off. I had a 2 year old with a block and it certainly went from was ok to screaming when we were at home. Not all can communicate their feelings.

Over my career there was a definite move to do more for the patient in front of the doctor at the expense of the many who had yet to meet the doctor. Personally I think it was a backward move.

I think there's also a wider question - utilitarian, if you like - placing a block, especially with catheter takes time - if you do 1 fewer patient on the list because of the that, is it that right/just on a societal basis?

I wonder what @dr-amit-pawa.bsky.social thinks of all this. Wouldn't be surprised if he told me he knew how to block tonsils.

I honestly think this is one of the smartest approaches to medical intervention uncertainty.

Dunning-Kruger curve, with and without AI. The era of Automated General Idiocracy is rising From: x.com/ryzerth/stat...

Fact 👇🏼

Good discussion here about what’s wrong with computing change from baseline in a parallel group randomized clinical trial. We need to abandon this practice. stats.stackexchange.com/questions/57... #Statistics #StatsSky

Doing shit statistics.

On another point. I have never understood why the moral distress of the bedside staff/code team is never acknowledged or discussed. We are considered automatons without feelings when nothing could be further from the truth.

The idea of this topic is to challenge outer ideas , so bets to chat with surgical colleagues who may follow our patients post op for longer than us, or to consider Day Patients? What happens when the block wears off and what you do?? #MedSkyDebate 4/4

@benmorton.bsky.social continues with “I know it’s sacrilegious to consider this, but does regional analgesia harm the patient. #MedSkyDebate 3/4

Consider a patient who has block with catheter for surgery, are pain free until removal of catheter and the increase in pain makes them feel they have regressed. If a single shot block is used they wake up with pain but it gets better every day.they have not regressed. 2/4