iiinova — where your digital self keeps evolving

Sophie Pilleron @spilleron.bsky.social · 03/12/2025

Please tell us if you succeed in explaining them why they can’t do that 🙏

010

Sophie Pilleron @spilleron.bsky.social · 02/10/2025

He will be coming next February. If you are around, please, come to the lecture.

130

Sophie Pilleron @spilleron.bsky.social · 12/07/2025

I am happy for you to present it :)

110

Sophie Pilleron @spilleron.bsky.social · 12/07/2025

What do you suggest when it is a common practice in some places ?

110

Sophie Pilleron @spilleron.bsky.social · 08/07/2025

No. In the paper I read, they use multiple multivariable logistic regressions on 1000s potential exposures of an outcome, to identify those which remain significant at a very low p-value threshold. They do another multivariable analysis, and after something about Mendelian randomisation.

110

Sophie Pilleron @spilleron.bsky.social · 08/07/2025

👋EpiBlueSky There's an increasing interest in exposome-related research using GWAS methods. I'm not an expert & not a fan of data-driven approaches, which look like p-fishing exercises. Does anyone know a resource which criticises these approaches or convinces me that they're good? 🙏

100

Reposted by Sophie Pilleron

SIOG NAHS @siog-nahs.bsky.social · 12/05/2025

Happy International Nurses Day! Today, we honour nurses in #GeriatricOncology — leaders in care, clinical expertise, & research. From assessments to symptom relief, coordination to end-of-life support, and advancing knowledge through research — your role is essential. 💙 Thank you for all you do.

media.tenor.com

a blue and white penguin says thank you

ALT: a blue and white penguin says thank you

021

Sophie Pilleron @spilleron.bsky.social · 09/03/2025

Not sure how to describe me if you were stupid 😬 You are very far from being stupid.

000

Sophie Pilleron @spilleron.bsky.social · 02/03/2025

Not here 🤔 It seems that it is more a technical issue than a political shut down.

010

Sophie Pilleron @spilleron.bsky.social · 02/03/2025

Not fully there yet ! But this was temporary ! This is a relief 🥲

120

Sophie Pilleron @spilleron.bsky.social · 02/03/2025

Pubmed is down ! That is unbelievable 😨

333

Sophie Pilleron @spilleron.bsky.social · 01/03/2025

Happy that our piece to present the @siogorg.bsky.social Methods Working Group I co-created with Dr Jennifer Lund is out 👇🏻 www.sciencedirect.com/science/arti...

sciencedirect.com

The International Society of Geriatric Oncology (SIOG) Methods Working Group

040

Sophie Pilleron @spilleron.bsky.social · 26/02/2025

Putting the finishing touches on a grant proposal—only to realise that a recently published paper is strikingly similar to what I had planned to propose!

media.tenor.com

a man is crying while looking at a cellphone

ALT: a man is crying while looking at a cellphone

010

Sophie Pilleron @spilleron.bsky.social · 26/02/2025

Happy to see this published! I had been thinking of writing something similar when I found the time. For all researchers working with quantitative data, this is a must-read 👇 Now I finally have a ref to attach to all my peer reviews! 😊https://www.neurology.org/doi/10.1212/WNL.0000000000210171

000

Sophie Pilleron @spilleron.bsky.social · 14/02/2025

I know about the residual confounding at T0. It was more about the Daggitty comment I was worried about. Thank you for your help. Hope you are doing well :)

010

Sophie Pilleron @spilleron.bsky.social · 14/02/2025

Actually, a bit more complicated than that. But basically observed only at T1. I tried with observed/unobserved etc still incorrectly adjusted.

000

Sophie Pilleron @spilleron.bsky.social · 14/02/2025

Observed at T1 but not observed at T0.

200

Sophie Pilleron @spilleron.bsky.social · 14/02/2025

#DAGsaddict 👋 Need help! Is there a way I can estimate the causal effect of sleep on cancer risk in the scenario below? DAGITTY tells me "Incorrectly adjusted" when adjusted on lifestyle factors at T-1 and T0. 🙏

211

Sophie Pilleron @spilleron.bsky.social · 10/02/2025

Grant rejected 😞 Have to wait to know why.

010

Sophie Pilleron @spilleron.bsky.social · 04/02/2025

On #WorldCancerDay, we highlight the unique cancer challenges older adults face & the need for better representation in research. Aligned w/ #UnitedbyUnique, ACADI advocates for person-centred oncology & equitable care. #CloseTheCareGap #GeriatricOncology 🙏 to India Pinker for designing this post!

030

Sophie Pilleron @spilleron.bsky.social · 01/02/2025

Dry January ✅ Now starting Dry February !

020

Reposted by Sophie Pilleron

Peter Tennant @pwgtennant.bsky.social · 31/01/2025

TIP: If you're planning a comparative effectiveness study in observational data, particularly a target trial emulation study, then please pre-register a study protocol! It makes a huge difference to how much readers will trust (or doubt) the analytical choices and results. #EpiSky

0222

Sophie Pilleron @spilleron.bsky.social · 30/01/2025

I like it :) Thanks for sharing.

010

Sophie Pilleron @spilleron.bsky.social · 27/01/2025

Seems super interesting but seems available in btw 3-7 months in France 🤪

000

Sophie Pilleron @spilleron.bsky.social · 22/01/2025

I do the same :) I hate having many tabs open, it makes me anxious 😬

020

Sophie Pilleron @spilleron.bsky.social · 17/01/2025

👋🏻Hello, It is said that there will be a shortage of #oncologists in the coming years. Would anyone have a reference for that statement?

010

Sophie Pilleron @spilleron.bsky.social · 16/01/2025

My brain does not want to understand what means marginal effect! The word marginal for me is so confusing.

130

Sophie Pilleron @spilleron.bsky.social · 09/01/2025

📣Excited to share the new #ACADI_Lux paper by Fei Liu, stemming from her 6-month internship, supervised by Dr India Pinker. ! 📄 What’s it about? A scoping review of online social support for #OlderAdults living with #Cancer. Check it out! 🎓✨ authors.elsevier.com/c/1kPY96hK~0...

media.tenor.com

a man in a clown costume is giving a thumbs up with the words check it out below him

ALT: a man in a clown costume is giving a thumbs up with the words check it out below him

100

Sophie Pilleron @spilleron.bsky.social · 05/01/2025

Small sample size does not introduce biases per se but produce imprecise estimates. Better to have an imprecise estimate than a biased one ;)

120

Sophie Pilleron @spilleron.bsky.social · 05/01/2025

I am not sure that providing an evidence from an analysis that is known to be biased is helpful in anyway to make informed decision. Sometimes we need to be humble and accept that we cannot analyse the relationship of interest because of impossible control for confounding.

141

Sophie Pilleron @spilleron.bsky.social · 05/01/2025

I am struggling now to justify why we are simply interested in a an association between X and Y without being interesting either in prediction or causality. What is the point to see whether X and Y are associated if it is not to know that X acts on Y or X predicts Y ?

131

Sophie Pilleron @spilleron.bsky.social · 19/12/2024

Hey 👋🏻 Has anyone a good resource on how to conduct a bibliometric analysis ? 🙏 #research #bibliometric

010

Sophie Pilleron @spilleron.bsky.social · 18/12/2024

The final #ACADI_Lux newsletter for 2024 is out 😄 Have a read 👇🏻 Thanks to India Pinker, PhD for writing and designing it 🙏🏻 Thanks to the Luxembourg Institute of Health and Luxembourg National Research Fund (FNR) for their support. Have a happy festive end of year 🎉 www.lih.lu/en/article/a...

lih.lu

ACADI's Holiday Newsletter » Luxembourg Institute of Health

Welcome back to ACADI’s holiday Newsletter! As the year draws to a close, we have some updates to share before wishing you all a very happy holiday season.

010

Reposted by Sophie Pilleron

Prof James B Kirkbride @kirkbride.bsky.social · 12/12/2024

Hello bsky. Good to be here. I'm hiring! Can't think of a better inaugural post than to share details of new 2-year research fellow post we're advertising for an outstanding epidemiogist to join my team to work on improving population mental health. tinyurl.com/28mlykr9 #CausalSky #EpiSky

tinyurl.com

UCL – University College London

UCL is consistently ranked as one of the top ten universities in the world (QS World University Rankings 2010-2022) and is No.2 in the UK for research power (Research Excellence Framework 2021).

13231

Reposted by Sophie Pilleron

Darren Dahly @statsepi.bsky.social · 12/12/2024

Predict != Usefully Predict. I'll bet that in the next 3 years we'll have a few hundred papers on micobiome + machine learning to predict long covid + some cluster analysis, and zero of them will lead to clinically useful insights.

78214