NEW PAPER out today in @BMJ_latest TRIPOD+AI: reporting recommendations for studies developing or validating prediction models for use in healthcare that use #machinelearning methods

bmj.com/content/385/bm…

#ArtificialIntelligence #AIstandards #OpenAccess

Please share 🙏

NEW PAPER in Journal of Clinical Epidemiology 'SPIN-PM: A consensus framework to evaluate the presence of spin in studies on prediction models'

--> tinyurl.com/4nwa4eb6

#machinelearning #spothehype #seethroughthehype #keepitreal

👇ONLY 10 PLACES LEFT to attend our UK Centre's ONLINE Publication School!

📅14-17 May 2024
⏲️12:30-17:00 (BST) GMT
🌎ONLINE- Zoom
✉️[email protected]

Sign up here--> tinyurl.com/3huarts4

NEW #OpenAccess paper (led by Paula Dhiman) in Journal of Clinical Epidemiology 'Prediction model protocols indicate better adherence to recommended guidelines for study conduct and reporting'

sciencedirect.com/science/articl…

Findings will feed into TRIPOD-P (guidance for protocols)

nature.com/articles/s4225…

Applications for our next ONLINE Publication School is now open! See post and link for more details!

NEW PAPER and some positive findings of practice in prediction model research

Prediction model protocols indicate better adherence to recommended guidelines for study conduct and reporting

shorturl.at/agJ58

All 3 papers ( #openaccess ) now out in the The BMJ on the **Evaluation of clinical prediction models**

part 1: tinyurl.com/3h4zfvfw

part 2: tinyurl.com/ynytrpmx

part 3: tinyurl.com/2cfc9uj3

#stats #MachineLearning #predictiveanalytics

NEW PAPER: Who knew that Avengers Endgame was actually about bootstrapping and instability?!

Head into the multiverse of madness and read about bootstrapping and instability from another perspective!

t.ly/DUA5V

*NEW PAPER superbly led by Brooke Levis *

'Risk of bias assessments in IPD meta-analyses of test accuracy & prediction models: a review shows improvements are needed' Journal of Clinical Epidemiology

- with guidance for undertaking RoB assessments in IPDMA of tests & models

sciencedirect.com/science/articl…

New paper: Sport Journals scored low on #OpenScience Much more work is needed, particularly code and data sharing
tinyurl.com/58pbtnyd
Gary Collins 🇪🇺 Dr Tom Hughes Franco Impellizzeri Patrick Ward Richard Riley (R²) Paula Dhiman stefan qz Brian Waterman, MD Dr. Anu M Räisänen Wake Forest Pitching Lab

NEW #openaccess paper in Journal of Clinical Epidemiology

#Openscience practices need substantial improvement in prognostic models in oncology using #machinelearning

-> tinyurl.com/468m8buw

Registration & protocols rare, data & code sharing uncommon, prediction models largely unavailable☹️

It 'should' be obvious but you can't use a prediction model when the predictors aren't known at the moment of prediction.

Here Paula Dhiman and I comment on one such prediction model...

--> tinyurl.com/5c56t6ef

#statstwitter #epitwitter

New paper: ‘How to check a simulation study’
now out in IJE

Ever run a simulation study and realised the results don’t make sense or you don’t believe them? Then this one is for you, because we’ve been there (too many times)!
academic.oup.com/ije/advance-ar…

We are hiring! 📣Are you a health economist with a passion for real world data, economic evaluations, decision making? Check this out!⬇️

ndorms.ox.ac.uk/jobs/research-…

#JobAlert #healtheconomics #research

Great opportunity to work with a great team!

In the July Issue of the Journal of Clinical Epidemiology we highlight several papers presenting issues on prediction models and prognostic tools David Tovey Argie Veroniki @RL_Hooper jclinepi.com/article/S0895-…

📢 NEWS: Joie Ensor releases 'pmvalsampsize' for Stata!

Calculates sample size needed for external validation of a prediction model - targets precise calibration, discrimination & clinical utility

tinyurl.com/2eac5whx
based on
tinyurl.com/5n7685rn
tinyurl.com/mtypcskk

NEW CORRESPONDENCE announcing the development of the TRIPOD-P reporting guideline - the protocol extension to TRIPODStatement and for studies planning to develop or validate a prediction model using statistical or machine learning methods

rdcu.be/dj17o

Xiao Liu

NEW PAPER: Sample size is not being considered when developing a prediction model

Only 8% of studies justified their sample size and unsurprisingly 73% didn’t meet their minimum required sample size (using Riley et al formulae) by a median of 75 events
tinyurl.com/yc8aw8uk

If you torture the statistician long enough they will confess they still don't fully understand your research question