Let's try something new: *7 days, 7 statistical misconceptions*. Over the next 7 days I'll post about 1 statistical misconception a day. Curious to hear your favorite #statsmisconceptions, so feel free to add you own to this thread

♕Deborah Mayo♕ Pawitan clearly didn't read our article and confuses P-values with "significance": His so-called "excellent article" falsely labels me and Amrhein as wanting to ban P-values when we called for nothing of the sort. I even wrote an article defending them: tandfonline.com/doi/pdf/10.108… .

"When a society loses its moral compass, it collapses from within." – G.K. Chesterton A thread uncovering the falling of the west. 1. The Celebration of Obesity

This is sad and sickening

I wrote the words below in April, about the UK left, and I'm retweeting them today for the benefit of some of the numbskulls in my mentions. Women like me, and there are a lot of us, aren't and never have been far-right. We simply want the left to wake the hell up, because we're

Rethinking energy & protein provision for critically ill patients 🔌 energy dose 🥩 protein dose 🍽️ learning points & proposed energy/protein provision 🚧 limitations and future directions Open access #FOAMcc #ICUrehab on Intensive Care Medicine A/Prof Emma Ridley 🔗 bit.ly/4eMfOj6

Lessons from robust RCTs Monitor daily -stable/recovering: ⬆️ E & P progressively - Increasing organ support lvl: regardless of ICU days, ⬇️ E & P to maintainence lvl -may not be the most optimal feeding algorithm, but a/w least harm (Pls read with supplement) A/Prof Emma Ridley

Classic paper established basis for valid indirect comparisons and ultimately network meta-analysis. Grasping simple principle that one gets indirect OR of A vs B by dividing the OR of A by the OR of B dissolves the mystery of indirect comparisons in NMA. pubmed.ncbi.nlm.nih.gov/9250266/

Classic study demonstrated how for measures of exercise performance the encouragement provided by the individual administering the test can have a marked effect on performance – the example, the six minute walk test. pubmed.ncbi.nlm.nih.gov/6505988/

Even more evidence for #ICUrehab in the new 2024 SCCM updated PADIS guidelines (below) Look for STRONG certainty re: effect of rehab on ICU-acquired weakness in this focused update to PADIS with much larger # of RCT in 2025 vs 2018 PADIS

First classic #JAMA paper demonstrating that randomized control trials (#RCTs) stopped early for benefit (truncated trials) systematically overestimate treatment effects, and when sample size small overestimates large. pubmed.ncbi.nlm.nih.gov/16264162/

Implementing early enteral nutrition in shock patients is challenging. This study proposed that enteral nutrition may only benefit those with quickly resolved shock. In patients with persistent shock, wait-and-delay may be not a bad option!

In this 11-minute video I use the example of our giant network #MetaAnalysis of all #diabetes treatments to explain the principles of #NMA. One commentary: This is a terrific overview that gets the point across in a remarkably straightforward way. #EBM #Medicine

BREAKING: U.S. Supreme Court rules that parents have the right to remove their children from public school lessons that include LGBTQ-themed storybooks.

For a long time, we were hesitating whether early drainage could be justified if a patient with necrotizing pancreatitis had stayed in the ICU for more than a week. Following the TIMING trial, we now know that the perceived benefits and concerns are NOT real.

Outstanding paper! Especially suited for non-statisticians who want to get started with relaxing linearity assumptions without resorting to dreaded categorization techniques ... @Statistics

🚀 Exciting news! The Acute Registry —a global initiative to evaluate ICU nutrition practices—is underway! Grateful for ~90 ICUs already on board 🌍💡 Want to join? Check out: [criticalcarenutrition.com/research/acute…](criticalcarenutrition.com/research/acute…) #CriticalCare #NutritionResearch #ICU

🔥 So exciting & important! The #AcuteRegistry will help transform ICU nutrition care worldwide—thrilled to see ~90 ICUs collaborating already 🌍👏 Let’s keep growing this! Retweeting 👇 #CriticalCare #NutritionResearch #health criticalcarenutrition.com/research/acute…

A lot of BS is being pushed to the scientific community in the form of dressing up inadequate design so that one can pretend to answer a question with inadequate data. Watch out.

It’s interesting that abandoning p-values and false discovery rates for 2nd generation p-values results in major reductions in decision errors: journals.plos.org/plosone/articl…