Stuart Miller

Sport Scientists bitching about bad #OlympicGames commentators knowledge of sport science... Let's apply the same logic when you create "novel statistical methods" or pontificate about statistics when you've had 3 or less classes on the topic your entire life.

Can we use Generative AI to support the clinician through training and in the real world? @QMULSportExMed @QMUL @Stuart_Miller @ManuAngioi @DrDylanM qmul.ac.uk/fmd/about/news…

It is 2024 and it is still common practice in the medical literature to put a bunch of covariates in a regression model for some bad health outcome and declare the significant associations as "risk factors" that people should worry about. Pseudo science

Personal reflection: "Clinical prediction models & the multiverse of madness" Thanks to @BMCMedicine for 'getting this' Many reviewers/Eds pushed for writing style & tone changes This thread delves into this & why we stuck to our original vision 1/n bmcmedicine.biomedcentral.com/articles/10.11…

@dsquintana I still don't think this really addresses the recommendations by Cohen and the weakness of using the general small/med/large thresholds. It assumes that the spread of ES in the literature are meaningful as small/med/large... they might all be small in terms of meaningfulness.

This package and paper was a huge effort from Bernt, who originally created this as part of his master thesis 👏 Let us know what you think!

In our new preprint led by Bernt Glaser, we introduce the ESDist R package for calculating field-specific effect size distributions for better understanding the magnitude of study results and for planning new studies. Let me walk you through it... 🧵 osf.io/preprints/psya…

We used to call this prognostic factor research Now being sold as data-driven, deep learning, AI, advanced covariate adjustment etc

Really interesting paper on Achilles rotation and impact on tendon strain - More rotation impacts tendon strain Frontiers | 3D Models Reveal the Influence of Achilles Subtendon Twist on Strain and Energy Storage frontiersin.org/articles/10.33…

This is the 3rd time this colleague gets this on their payslip, for withholding 13 scripts. Management threatens 2 more payslips like this. MAB is over, taking this money can't shorten it. This is not proportionate but punitive; not for the sake of students but to punish staff.

Just a reminder for anyone measuring ultrasound echo intensity (EI) as an indicator of 'muscle quality' or adiposity - fascicle angles affect the sound echo intensity, so any pennation changes with ageing, exercise training or disuse will influence EI...europepmc.org/article/med/36…

The word *risk factor* is a risk factor for a completely uninformative data analysis

In Sports Medicine and Sport Science, we continue to ask the "wrong questions" and interpret those nonsense "answers" in ways that don't make sense. You often don't care if there is a relationship between X and Y. You want to know if changing X causes a change in Y.

Academic publishing

Congratulations to all our iBsc 2022-23 Sports Ex & Med students!! Some great research outputs shared throughout two presentation days @QMULSportExMed @QMULBartsTheLon #sportsmedicine #iBSc

This weekend I'm taking on Endure 24 solo. Like when I ran the marathon a few years back, I want to take this opportunity to get you thinking about all things mental health, and if you are able to (and want to) @Rethink_ is a great charity to donate to. justgiving.com/fundraising/st…

Great job @q8mpt really doing a deep dive in to possible local mechanisms for treatment effects.

Put a bunch of covariates you happen to have measured in a regression model to predict an outcome and claim it's a very "explainable" model because you have multivariable regression coefficients to look at

Mandatory periodic reminder to this work of art

We look forward to a world that prioritises mental health for 52 weeks a year, not just one. #MentalHealthAwarenessWeek

Our new paper in @FrontSportsAL is now available. We observed how a 2-week neuromuscular exercise programme affected biopsychosocial measures in people with PFP. Unlike what has been observed in healthy populations, little changed in this cohort. frontiersin.org/articles/10.33…