Pleased to have presented my work about the use of pseudo-observations in the development and validation of prognostic models for competing risks at ISCB45! Thanks to all that attended and asked questions. Feel free to reach out if you have any questions! #iscb45

My poster at #iscb45 on adding pregnancy complications to 10 year risk of CVD prediction. Got a chance to meet and take Ben Van Caster Ben Van Calster through my poster as well. #mumpredict

Developing a prediction model? Ever wondered how to target sample size to improve model fairness and precision of risk estimates? Check out our new pre-print & software package pmstabilityss arxiv.org/abs/2407.09293

Prepare to meet Joy! Nicola Coughlan sent a special message to attendees of Hall H at San Diego Comic-Con... and a first glimpse of #DoctorWho's upcoming festive special! 🎄

Great and much needed letter 👇

** NEW PAPER ** "Calculating the power of a planned individual participant data meta-analysis to examine prognostic factor effects for a binary outcome" Led by Rebecca Whittle Calculate power of IPDMA to investigate (un)adjusted prognostic factor effects onlinelibrary.wiley.com/doi/10.1002/jr…

New presentation: tips for optimizing collaborations of biostatisticians and non-biostatistician biomedical researchers: fharrell.com/talk/collab #Statistics #MedicalResearch #clinicaltrials #clinical Vanderbilt Department of Biostatistics

Free Textbook on Applied Regression and Causal Inference statmodeling.stat.columbia.edu/2024/07/30/fre…

I'm looking for staunchly Bayesian takes on calibration. My current view is calibration is meaningful to a Bayesian as a signal of how well they're using available info, but they don't value it in-and-of-itself like a Frequentist does. But curious if I'm overlooking anything.

Out today from Doctor Who Magazine, a special edition devoted to missing Doctor Who. Includes CG reconstructions of sets from TWO missing episodes, done by clever Gav Rymill 🤖 and scribbled on by Rhys and me.

Another JOB in our team but sadly X insists that University of Bristol job advert links are potentially dangerous!😂 For anyone interested in a part time (21 hrs/week) post doc role working on an individual participant data (IPD) meta-analysis of diagnostic test accuracy... (1/2)

Happy to present this NEW PAPER of the The STRATOS initiative on 'Regression without regrets: Initial data analysis is a prerequisite for multivariable regression' bmcmedresmethodol.biomedcentral.com/articles/10.11… Frank Harrell Marianne Huebner #stratosida

🧬Join us for a 3-day in-person course on Mendelian randomisation at UCL from 24th-26th September. This course introduces some of the theory and most widely used tools for MR, and how to apply these methods to real research questions. tinyurl.com/7rrv8brf

The Better Methods for Better Research (BMBR) Programme, a joint partnership between Medical Research Council and NIHR, will now be managed by NIHR with an increased budget. Find out more: nihr.ac.uk/news/nihr-take…

In our NEW Journal of Clinical Epidemiology paper we present practical guidance & recommendations for multinomial prediction models 📈👩🏻‍💻💡 Glen Martin Ben Van Calster Jamie Sergeant doi.org/10.1016/j.jcli…

Looking forward to #RSS2024 - speaking on Wednesday in early session ‘Topics on predictive modelling in med stats’ Title: “Sample size calculations for developing a clinical prediction model: an approach to target precise individual-level predictions for decision making”

Interested in developing a clinical prediction model? Check out our new paper 🥳🥳🥳 in the The BMJ where we present a step-by-step guide bmj.com/content/386/bm… Together with Matthias Egger Georgia Salanti Konstantina Chalkou Thomas Debray. Special thanks to Richard Riley (R²) for his help!

This morning at 8:50am

Preprint for our new sample size work for developing a prediction model in healthcare arxiv.org/abs/2407.09293