“FO”? Sounds like there’s a possibility you’re going to turn up…

I’m at EuroCIM for the first time and
a) lovely community with lots of familiar faces & nice informality
b) the talks are just as technical as the abstracts imply

A funny thing happened a few times today: speakers slipped into a reverie while pondering some dense notation on a slide.

It’s a relief to be able to set AI to work on boring tasks so I can spend my time on important things like opening blinds, MFA, and proving I am still human.

A beginner’s guide to hot-desking etiquette:
A) Pick desk by window
B) Close blinds

There was a long delay between Fleming et al’s paper and our comments… I suspect there will be a similar one before we get to see their response. Not sure if I’m looking forward to reading that or not!
6/

Commentary on Fleming et al. “A Perspective on the Appropriate Implementation of ICH E9(R1) Addendum Strategies for Handling Intercurrent Events” Click on the article title to read more.

The third comes from the Estimands Implementation Working Group.
‘Commentary on Fleming et al. “A Perspective on the Appropriate Implementation of ICH E9(R1) Addendum Strategies for Handling Intercurrent Events”’
by Oliver Keene, David Wright and Crissie Fletcher
5/
doi.org/10.1002/sim....

What Makes an Estimand Useful? Guidance on the Choice of Intercurrent Event Strategies While the use of estimands in randomized trials is increasing, there is little guidance on which intercurrent event strategies should be used. The article by Fleming et al. seeks to address this gap....

There’s a very nice commentary on the tension between relevance and reliability.
‘What Makes an Estimand Useful? Guidance on the Choice of Intercurrent Event Strategies’
By @brennankahan.bsky.social Fan Li, Michael Harhay & @suziecro.bsky.social
4/
doi.org/10.1002/sim....

Unfortch, their message seems to be roughly ‘ITT or bust’.

This prompted three commentaries. Ours tries to give a different perspective that emphasises the flow of reasoning (figure 1 particularly), that starts with estimand and moves forward, but may get stuck and need to go backwards.
3/

A Perspective on the Appropriate Implementation of ICH E9(R1) Addendum Strategies for Handling Intercurrent Events In randomized clinical trials, stopping study medication, use of rescue treatment, and other intercurrent events can complicate interpretation of results. The ICH E9(R1) Addendum on estimands stimula...

This comments on a paper that appeared last year, ‘A Perspective on the Appropriate Implementation of ICH E9(R1) Addendum Strategies for Handling Intercurrent Events’.
The aim – ‘Scientifically Rigorous Causal Inference on Meaningful Estimands’ – is good.
2/
doi.org/10.1002/sim....

A Causal Perspective on “Appropriate Implementation of ICH E9(R1) Addendum Strategies” (Comment on Fleming et al.) This commentary offers perspectives on delivering “rigorous causal inference on meaningful estimands” that differ from the opinions recently shared by Fleming et al. We (1) depict a more robust pathw...

New commentary in Statistics in Medicine:
‘A Causal Perspective on “Appropriate Implementation of ICH E9(R1) Addendum Strategies” (Comment on Fleming et al.)’
By me, Alex Ocampo, Jesper Madsen, Hege Michiels, Sanne Roels.
1/
doi.org/10.1002/sim....

Non-inferiority has a weird relationship with estimands. The whole PP convention seems to be a reaction to “what if people use ITT to cheat?”
Seems more relevant to begin with “What effect do you want to learn in this TB trial?”

Because you’re trying to decide which you need measured?

For anyone who is doing statistical simulation work, I honestly cannot recommend @timpmorris.bsky.social's elegant 2019 manuscript on ADEMP framework enough. Outlining a structured approach to planning sim studies, w/in-depth examples+how to display Monte Carlo SEs. #statssky#statistics #academicsky

Thank you @dszlosek.bsky.social, this has made my day!

Not that I am in any danger of this

Note to self:
I never ever want to present an RSS discussion paper. The experience of ‘discussions’ like these would drive me to the bottle.

+1

A colleague just wrote “Small Ns” in a chat and it reminded me of the Rabois fallacy, and I want to remind you too so that you get to experience that wistful chuckle

%%title%% %%sep%% Maastricht University Press Open Access book by Luc Smits, Sander van Kuijk, and Laure Wynants: This open-access textbook offers a practical and comprehensive guide to developing, validating, and implementing clinical prediction...

✨ New open-access book✨
I’ve seen how powerful prediction models can be, but also how often they fall short. We wrote a book, covering not just development, but also when models are needed, and how to ensure real-world impact.
www.maastrichtuniversitypress.nl/cpm

HELLO! 😍

Obviously we had this conversation with ADEMP

Great, you managed to get “INSPIRED” out of those bullets… now, what were the bullets again?

Laboriously-constructed acronyms:
- Good for trials because it’s just about having a shortcut name.
- Bad where people need to remember the elements.

I simply refuse to believe this isn’t a Chris Morris sketch:

Future work: “We used all the same methods as the older work, for consistency”

Grand Taxonomy of Rap Names Here are 282 sobriquets from the world of rap music, arranged according to semantics. Version 1.5 contains 16 additional rappers, including KRS-One, Mos Def and Tech N9ne.

Sort of reminds me of this
popchart.co/products/gra...

🙌

TBF that bit is ok. A certain form of MMRM is equivalent to a certain MI, without the Monte Carlo error of MI. Calling MI (without any further info) a sensitivity analysis for this is wrong.

Email subject screenshot: “Reminder: Invitation to join the Metascience Novelty Indicators Challenge”

I assume lots of you are also receiving this. Somehow the entire message comes off as dripping with contempt.
Hooold!

Stumbled across this podcast on clinical trials today and enjoyed it. I like skilled interviews talking to experts and this is that. cc @andygarrett.bsky.social @timpmorris.bsky.social

open.spotify.com/show/1FfO4fr...

I love when someone quotes a good bit of ancient wisdom