Gary Collins

Welcome! You are invited to join a webinar: Research Methods Academy: Seminar - The Importance of Sample Size for the Development and Evaluation of Clinical Prediction Models. After registering, you w... Led by Professor Richard Riley Join this Online Seminar on clinical prediction models estimate individual risk for outcomes from disease. Many studies lack rigorous design; this session will focus o...

Giving a talk today which anyone can watch online: 1pm - 2pm
"The Importance of Sample Size for the Development and Evaluation of Clinical Prediction Models"
- intended for a broad audience (anyone doing or using prediction model research)
imperial-ac-uk.zoom.us/webinar/regi...

Peer review of prediction model studies in oncology needs improvement: a systematic review of open peer review reports from BMC journals - Journal of Clinical Epidemiology www.jclinepi.com/article/S089...

Drummond Rennie (1936-2025), in his own words Drummond Rennie I first became aware of the work of Drummond Rennie almost by accident: By borrowing his office. It was the summer of 1997, and as a rising fourth-year medical student, I was spendi…

Saddened to hear of the passing of Drummond Rennie, a quite influential figure. retractionwatch.com/2025/09/15/d...

NEW PAPER "The STARD-AI reporting guideline for diagnostic accuracy studies using #artificialintelligence"

--> www.nature.com/articles/s41...

#machinelearning #MLsky #statssky #digitalhealth #transparency

INSPECT-SR: a tool for assessing trustworthiness of randomised controlled trials The integrity of evidence synthesis is threatened by problematic randomised controlled trials (RCTs). These are RCTs where there are serious concerns about the trustworthiness of the data or findings....

www.medrxiv.org/content/10.1...

INSPECT-SR: A tool for assessing trustworthiness of randomised controlled trials.

Biostatistician - Cancer Research UK Clinical Trials Unit - 105965 - Grade 7 This post is for a Biostatistician to be the main statistical scientist for a methodological project on improving statistical methods for the design and analysis of clinical trials in paediatric cance...

Job available for a statistician to work on a clinical trials methodology project on improving statistical methods for paediatric cancer ttrials.

Cancer Research UK Clinical Trials Unit, University of Birmingham

Closes 22 September

edzz.fa.em3.oraclecloud.com/hcmUI/Candid...

Figure. SPIRIT 2025 Diagram of the Schedule of Enrollment, Interventions, and Assessments

📌 The SPIRIT 2025 statement updates the guidelines for randomized trial protocols to enhance their completeness, transparency, and consistency, benefiting stakeholders involved in clinical research.

ja.ma/4fzMVZS

TRIPOD+AI and CHART target different studies.

TRIPOD+AI is for studies describing the development/evaluation of non-generative AI models for diagnosis/prognosis.

CHART is for studies evaluating generative AI driven chatbots that summarise clinical evidence and providing health advice.

Reporting guideline for chatbot health advice studies: the Chatbot Assessment Reporting Tool (CHART) statement The Chatbot Assessment Reporting Tool (CHART) is a reporting guideline developed to provide reporting recommendations for studies evaluating the performance of generative artificial intelligence (AI)-...

Reporting guideline for chatbot health advice studies: the Chatbot Assessment Reporting Tool (CHART) statement bmjmedicine.bmj.com/content/4/1/... #llms #chatbots #bioinformatics #healthcare

UKRI award rate halves in seven years as grant applications double - Research Professional News Trends across research councils prompt warning that system is “overstretched or broken”

UKRI award rate halves in seven years as grant applications double. www.researchprofessionalnews.com/rr-news-uk-r...

NEW PAPER: the CHART reporting guideline for studies evaluating generative AI models like #ChatGPT & other #LLMs for providing health advice

--> bmjmedicine.bmj.com/content/4/1/...

The explanation & elaboration paper

--> www.bmj.com/content/390/...

#genAI #chatbots #healthadvice #reporting

New paper in BMJ (Research Methods & Reporting):

“Dealing with continuous variables and modelling non-linear associations in healthcare data: practical guide”

led by Pedro Lopez-Ayala & Tobias Zimmermann with & myself

www.bmj.com/content/390/...

Self tests

Many high street health tests are unfit for purpose and need greater regulation, warn researchers in two studies.

These studies highlight issues over accuracy and suitability for public use, they say
www.bmj.com/content/390/...
www.bmj.com/content/390/...

Cover picture with blog title & subtitle, and results graph in the background

Post just up: Is multiple imputation making up information?

tldr: no.

Includes a cheeky simulation study to demonstrate the point.
open.substack.com/pub/tpmorris...

Linear predictor plot for three modelling approaches to analyse continuous variables in a case study of cerebrospinal fluid glucose and acute bacterial meningitis

This article provides an overview of the current state of handling continuous variables in healthcare research.

It discusses the potential limitations of assuming a linear relationship between independent and dependent variables
www.bmj.com/content/390/...

NEW PAPER in @bmj.com "Dealing with continuous variables and modelling non-linear associations in healthcare data: practical guide"

--> www.bmj.com/content/390/...

#methodologymatters #StatsSky #EpiSky

**NEW OPEN-ACCESS PAPER**

“A decomposition of Fisher’s information to inform sample size for developing or updating fair and precise clinical prediction models for individual risk—part 1: binary outcomes"

- published in Diagnostic & Prognostic Research
- implemented in pmstabilityss

1/n

NEW PAPER in @jclinepi.bsky.social looking at the content of open peer-review reports of RCTs in the field of oncology

www.jclinepi.com/article/S089...

Peer review reports [...] were short (median 276 words), superficial, and rarely constructive

#metascience #peerreview #RCTs

Very nice summary of the confusing and often ill-posed metrics.

* New preprint led by Joao Matos & @gscollins.bsky.social

"Critical Appraisal of Fairness Metrics in Clinical Predictive AI"

- Important, rapidly growing area
- But confusion exists
- 62 fairness metrics identified so far
- Better standards & metrics needed for healthcare
arxiv.org/abs/2506.17035

NEW PREPRINT "Critical Appraisal of Fairness Metrics in Clinical Predictive AI"

-> arxiv.org/abs/2506.17035

We identified 62 fairness metrics (and growing) - unsurprisingly it's all a bit of a mess...with most metrics not fit for purpose

#predictiveAI #fairness #machinelearning #StatsSky #MLSky

And the exciting announcement! The 9th WCRI is set to take place in Vancouver, Canada, in 2026! Look forward to seeing you all there!

For researchers with limited background in statistics, we have written this gentle introduction to simulation studies (with @timpmorris.bsky.social and other people from the STRATOS initiative):
bmjopen.bmj.com/content/10/1...

REF 2029: splitting staff from outputs ‘distressing’ in jobs crisis - Research Professional News Decision to hold fast on portability prompts criticism over potential impact on early career researchers

REF 2029: splitting staff from outputs ‘distressing’ in jobs crisis.

Decision to hold fast on portability prompts criticism over potential impact on early career researchers.

One of our most-read articles this week.

www.researchprofessionalnews.com/rr-news-uk-r...

A brief history of medical statistics and its impact on reproducibility. You have to admit that medical research is a strange bird.

Medical research is special because it's too often led by high-status novices that don't know they are novices. The worst of these are also arrogant and insecure. Literally everyone else involved in the research has to work around these people.

Good morning! 😀

(ICYMI)

CONSORT 2025 Statement This Special Communication introduces the updated Consolidated Standards of Reporting Trials (CONSORT) 2025 statement, which includes substantive and structural changes to account for recent methodolo...

The CONSORT 2025 Statement is here. Essential reading for those involved in the reporting & reviewing of randomized trials.

We're proud to see so many #UKCTUNetwork units & individual researchers contributing to this crucial update!

🔗 tinyurl.com/yc3cbh33

#CONSORT2025 #ResearchTransparency

At #CCR25, we saw some trials use a #CompositeOutcome (an endpoint that combines more than one outcome into a single measure) & also saw some of the problems these can throw up. I have discussed these before but thought it would be useful to revisit 1/9
#MethodologyMonday #120

CONSORT 2025 Statement Updated Guideline for Reporting Randomized Trials

Authors, editors, reviewers, and other potential users should use CONSORT 2025 when writing and evaluating manuscripts of randomized trials to ensure that trial reports are clear and transparent.

Learn more in this Special Communication.

https://ja.ma/3TlnKQe

How long is acceptable for a journal to make a decision on whether to even review your article? We've got an article under submission for 5 weeks now, and still 'With Editor' for deciding whether to send out for review

Surely too long (journal even promises 'quick turnaround for initial decision')