I just responded to 67 reviewer comments on a paper. What do I win??? ✨

Nailed it.

Spent a week reading 100 random SRs of prediction models in medicine published >2025. Takeaways: - Review mills, the same authors leading 10-15 SRs in 1 year - Q1 journals publishing terrible SRs with nonsense PRISMA diagrams - The #1 outlet for SRs of predictive models is Cureus :( Great! 💣

Read this for the thread below on why it's misleading and the Lancet should feel bad

The Lancet@thelancet.com · 6mo ago

AI-enabled stethoscopes show promise for improving diagnosis of cardiovascular conditions, UK trial finds. Explore the research: spkl.io/63326AsjJ8

"Participants with an ED discharge outcome of death were excluded from the final analysis as this impacted on their re-presentation rate." Hmmmmmmm..... #statsky #biostats

Just a normal week eh #academia #momademia

Hello #statsky folks I had a baby but I'm back. What did I miss? 😅

New #preprint with @nicolewhite.bsky.social and @tayacollyer.bsky.social. We examined if Australian ethics committees use qualified statistical expertise. Our results revealed a large variation in practice, with some committees using qualified statisticians […] (1 of 4) osf.io/preprints/os...

At the stage of data analysis where I don't even know how many people are in the cohort, nothing is consistent and nothing adds up 👍 (Been around long enough to know this happens on every single project with big, messy, administrative data)

a man in a suit is holding a jacket and a cigarette .

ALT: a man in a suit is holding a jacket and a cigarette .

media.tenor.com

Me trying to get it all done before maternity leave: #statsky #episky

Ok this one is a lot 👀 #statsky #biostats

@statanon.bsky.social · 2y ago

Was asked to give stats advice on an RCT of stem cells from placental tissue. I sat in the meeting, 30 weeks pregnant, while (male) PIs discussed logistics of getting my donated placenta across state lines so it could be included. No one asked my thoughts. I have never felt so invisible. #episky

In clinical research, you will often receive feedback on study design, stats, and/or data analysis from an editor or reviewer that is simply wrong. Here is a list of common "statistical myths" and references you can use to push back. discourse.datamethods.org/t/reference-...

Reference Collection to push back against "Common Statistical Myths"

Note: This topic is a wiki, meaning that this main body of the topic can be edited by others. Use the Reply button only to post questions or comments about material contained in the body, or to sugge...

discourse.datamethods.org

Clinician on Friday: I have 20 patients, 5 have a condition and 15 do not. How do I get a p-value showing that my data is significant? Me: What is your hypothesis? Clinician: I don't have one. How do I get a p-value? #statsky #biostats #episky

Ok I don't know what is happening but I'm grateful for all the new followers! Get ready for some cheerful biostats snark in your feed! 🏆

This is so important. Routine, unthinking adjustment for race is a really bad habit. In peer review I have started asking authors to outline the explicit hypothesis about race, or remove race from multivariable analyses. (So far everyone has responded by removing race) #episky #statsky

Luisa N. Borrell@lborrell.bsky.social · 2y ago

Race and lung function equation: The use of the race-neutral equation could increase annual disability payments by 17%, or $1.1 billion, for Black veterans and drop by about 1%, or > $500 million, for White veterans. #Racism undark.org/2024/09/24/t... www.nejm.org/doi/pdf/10.1...

Crikey this is so true. I went too low with a recent paper and the review was literal gibberish. Parsing and responding to gibberish is exhausting... #academia

Tatyana Deryugina@tderyugina.bsky.social · 2y ago

#2: Don’t assume a lower-ranked journal will be less work. The bar for an R&R may be lower, but the number and difficulty of things you will be asked to do may not be.

Real conversation from biostat consult (outcome was measured on discharge from hospital): Biostat: 'I notice you excluded the participants who died?' Investigator: 'Yes, because there was no outcome' Biostat: 'I would argue that there is an outcome...' #statsky #biostats

Reviewing a study which treated death and stenting as equivalent in survival analysis. Just want to share that with the group. 💣 #statsky

Fun Thursday Problem: Cohort N=3960, but dataset has 4,503 unique IDs in it. (I set up the cohort and dataset from a starting point of 551,176 presentations. I have no one to blame but myself 😜) #statsky

I feel we need a #biostats feed... for bringing together the crew from Ye Olde StatsTwitter. I do not know how to make a feed, but if no one else does it I will try next week! #statsky #academicsky

This week I have been called Tara, Tanya and Tayla in email correspondence by other academics. It is a vibe! #academicsky

Statisticians! Please use hashtags like #stats, so your work or question or idea appears in the StatsSky feed and can be found by people interested in that topic! And please promote other people's work there too, especially if they aren't typically represented in science feeds.

"What's the aim of your analysis? What is your research question" Clinician who fit 14 versions of the same regression and came to me for help: "This wasn't a part of the original project. I just wanted to see if anything was significant. I don't really have an aim." #biostatistics #academia

"I did an analysis for the director of a clinical group at my hospital and they published it without naming me, or allowing me to see/edit the manuscript. They cheerfully told me it had been published when I ran into them in the carpark about 3 months later."

👉 (bio)statistician with an axe to grind? 👉 Want to anonymously vent about a tough experience? 👉Was the grant due tomorrow? 🏆statanon.bsky.social is live! Email your gripes to biostatanon@gmail.com or send a DM #statistics #academia

Evergreen post 👇

Taya Collyer@tayacollyer.bsky.social · 3y ago

Me in every stats consult 👇

Got a pub underway which has been rejected 4 times. Good suggestions from #4 which require a lot of extra reading and analysis. I'm going do it, cut a section, and submit to #5 Sometimes this is just how the work goes! I am outside my main area of expertise and grateful for the input 💪

The best part about publishing my recent qual paper was when the editor at Journal #4 refused to share the reviewer comments which underpinned their decision to reject. All I got was a ~50 word extract from one of the reviews. Science! 🥂 #statsky #academicsky

One of my most frequent comments in statistical peer review. I can tell when you have no idea what the null hypothesis is! #statsky

Hi friends. I'm looking for someone with solid medical stats and/or health data science skills to work with me on development of clinical prediction models for diagnosis and prognosis. It's a Postdoctoral Researcher position and is full time for 2 years. Search ref: 074011 ore.ucc.ie