Lightnews — Scholar-powered news

Sam Leighton @samleighton.bsky.social · 6d

Cool that makes sense. Cheers

Sam Leighton @samleighton.bsky.social · 6d

Is this still as problematic if they regress out the weight at baseline? In this paper they don't seem to do that, rather the control of recent weight change which isn't the same thing.

1

Sam Leighton @samleighton.bsky.social · 26d

Dynamic prediction with prediction of counterfactuals for individualised decisions. Really cool.

JAMA @jama.com · 27d

Among preterm infants with severe thrombocytopenia, this modeling study found substantial variation among individuals in predicted benefits and harms of prophylactic platelet transfusion based on their current clinical characteristics.

ja.ma/3KmMupS

Figure 3, Distribution of Risk Estimates by Transfusion Strategy: A graph comparing observed and predicted risks of major bleeding or death with and without prophylaxis, including discrimination and calibration data.

Reposted by Sam Leighton

Georgia Tomova @georgiatomova.bsky.social · Aug 31

A fantastic teaching example of a predictor being misinterpreted as having a causal effect.

#CausalSky #EpiSky

First week 'critical' to avoid children missing school later, parents told

Data suggests more than half of children who miss a day at the start of term become persistently absent.

www.bbc.com

5 19 110

Sam Leighton @samleighton.bsky.social · Aug 19

I also have this issue - no moderation despite preprints being available for months. It will significantly reduce the utility of psyarxiv if moderation takes the same time as peer review for an actual journal. There will be little point in submitting to preprint at psyarxiv if this is the case.

1 6

Sam Leighton @samleighton.bsky.social · May 21

New causal prediction feasibility study preprint with I Lam Leong, @dmachlanski.bsky.social @benjaminperry.bsky.social @stsaftaris.bsky.social Fani Deligianni, Stephen M Lawrie and @rajkd.bsky.social

Sam Leighton @samleighton.bsky.social · May 21

Check out our new preprint: Antipsychotic-induced weight gain in psychosis – a causal mediation analysis and feasibility study of casual actionable prediction model development using counterfactuals to target obesity osf.io/preprints/ps... #CausalSky

OSF

osf.io

1 1

Sam Leighton @samleighton.bsky.social · May 21

Check out our new preprint: Antipsychotic-induced weight gain in psychosis – a causal mediation analysis and feasibility study of casual actionable prediction model development using counterfactuals to target obesity osf.io/preprints/ps... #CausalSky

OSF

osf.io

3

Sam Leighton @samleighton.bsky.social · May 17

Mazel Tov

1 1

Reposted by Sam Leighton

Rajeev KD @rajkd.bsky.social · May 2

link.springer.com/article/10.1...

Ethnic Bias in Prediction and Decision Making Algorithms in Precision Psychiatry: Challenges in a Shrinking World - Journal of Psychosocial Rehabilitation and Mental Health

This article examines ethnic bias in predictive algorithms and decision-making systems within precision psychiatry. While these models aim to provide individualised outcomes and improve healthcare, th...

link.springer.com

1 1

Reposted by Sam Leighton

Sam Leighton @samleighton.bsky.social · Feb 24

Check out the new preprint with @rajkd.bsky.social @stsaftaris.bsky.social and Fani Deligianni
A fallacy at the heart of precision medicine - are current clinical prediction models actionable? osf.io/preprints/ps... #CausalSky

OSF

osf.io

1 2

Sam Leighton @samleighton.bsky.social · Feb 26

Thanks. I would love to attend but unfortunately have clinical duties at the time.

Sam Leighton @samleighton.bsky.social · Feb 25

Will this be recorded and available after the webinar?

1

Sam Leighton @samleighton.bsky.social · Feb 24

Check out the new preprint with @rajkd.bsky.social @stsaftaris.bsky.social and Fani Deligianni
A fallacy at the heart of precision medicine - are current clinical prediction models actionable? osf.io/preprints/ps... #CausalSky

OSF

osf.io

1 2

Reposted by Sam Leighton

Gary Collins @gscollins.bsky.social · Feb 14

NEW PAPER in @bmj.com where we discuss the rationale, challenges, and approaches to providing uncertainty
of risk estimates from clinical prediction models

www.bmj.com/content/388/...

#StatsSky #MLSky #EpiSky #AI #machinelearning

14 39

Sam Leighton @samleighton.bsky.social · Jan 31

Me and @rajkd.bsky.social published a paper on precision psychiatry www.cambridge.org/core/journal...

Precision psychiatry: thinking beyond simple prediction models – enhancing causal predictions | The British Journal of Psychiatry | Cambridge Core

Precision psychiatry: thinking beyond simple prediction models – enhancing causal predictions

www.cambridge.org

3 7