Matteo Quartagno

A card with a quote from Matteo Quartagno. It reads “I’m proud to have contributed to the development of innovative trial designs that reduce the burden on patients and healthcare systems while helping us identify better treatment strategies more efficiently. Approaches such as MAMS, PRACTical and ROCI have shown how trials can be more adaptive, resilient and impactful. What makes this even more rewarding is how receptive and engaged everyone involved in our trials – from clinicians and trial managers to data managers and research nurses – has been in embracing these innovations. Their commitment has turned methodology into practice, and I’m proud that together we are shaping research that will ultimately improve care for patients.”

It’s time for another #ProudestOf Wednesday!

Today, we’re spotlighting Matteo Quartagno (@matteoq21.bsky.social), Principal Research Fellow, who’s advancing the field with innovative trial designs as part of our #Methodology research.

Here’s what Matteo is #ProudestOf from his work at the Unit 👇

A comparison of different population-level summary measures for randomised trials with time-to-event outcomes, with a focus on non-inferiority trials - PubMed When the hazards are likely to be approximately proportional, reflecting this in the analysis can lead to large gains in power for difference in restricted mean survival time and difference in surviva...

This work was amazingly led by Lana Broer, and here is the paper with the theoretical results it was based on:

pubmed.ncbi.nlm.nih.gov/37337728/

Usually, the life of a methodologist is*:

- Develop fancy method to solve problem;
- Show through simulations that it makes a huge difference;
- Show application in which it makes no difference at all;

So, great for once to see empirical results matching our expectations!

* cit. James Carpenter

Summary measures in non-inferiority clinical trials with a time-to-event outcome: an empirical comparison of power - BMC Medical Research Methodology Background Time-to-event data is commonly used in non-inferiority clinical trials. While the hazard ratio is a popular summary measure in this context, the difference in restricted mean survival time ...

Hot off the press:
bmcmedresmethodol.biomedcentral.com/articles/10....

In this we confirm empirically some theoretical results about using different summary measures in non-inferiority trials with survival outcomes. tldr: RMST beats HR.

We still have few places available for our short course on sample size calculations. If interested, book your spot by the end of this week!

Reference-Based Multiple Imputation for Longitudinal Binary Data Reference-based multiple imputation provides a practical information anchored tool for inferences about the treatment effect for a treatment policy estimand with a longitudinal binary outcome. The latent multivariate normal model is the preferred implementation.

This new paper explores the best way to implement reference-based multiple imputation for longitudinal binary data in #ClinicalTrials where data are missing after participants deviate from their assigned treatment.

Read now👇
https://buff.ly/4jJtGOW
@matteoq21.bsky.social @suziecro.bsky.social

I think what Darren is suggesting is that you open an OnlyFans account, and use it to post exclusively Stata related content. I am sure it will be a huge success!

Multiple Imputation for Longitudinal Data: A Tutorial Longitudinal studies are frequently used in medical research and involve collecting repeated measures on individuals over time. Observations from the same individual are invariably correlated and thu....

Hot off the press! 📣📣In this tutorial we illustrate available multiple imputation approaches for handling longitudinal data including when they are clustered within higher level clusters. A reproducible example with R and Stata code provided! #OpenAccess

onlinelibrary.wiley.com/doi/10.1002/...

Reference‐Based Multiple Imputation for Longitudinal Binary Data Introduction In clinical trials, a treatment policy strategy is often used to handle treatment nonadherence. However, estimation in this context is complicated when data are missing after treatment .....

📢New publication: How to implement Reference-based multiple imputation for longitudinal binary data, lead by
@suziecro.bsky.social with @matteoq21.bsky.social Ian White & James Carpenter onlinelibrary.wiley.com/doi/10.1002/...

🤣 Another STaTa prize coming to the CTU soon...!

BTW, reading this list it just occurred to me why so many people think it is called STATA: it's the logo, that clearly has capital Ts in it!
SO perhaps we should really call it STaTa. 😜

Amazing, thanks, I echo Tim's sentiment! :)

A vanilla KMunicate-style Kaplan–Meier graph from {kmunicate}. Key is the extended risk table showing the status of participants over time.

Stata users: if you want to make ‘kmunicate’-style Kaplan–Meier graphs, I’ve just released a Stata package on GitHub.

To get it:
. net from raw.githubusercontent.com/tpmorris/kmu...
User feedback welcome!
1/

Ian White, Professor of Statistical Methods for Medicine at the MRC Clinical Trials Unit at UCL, has been awarded the 2024 Stata Journal Editor's Prize for his outstanding contributions to the journal.

Congratulations Ian! 🎉 Read more about the award 👉 https://buff.ly/3BGCUKU

NEW PREPRINT

A detailed overview of 32 popular predictive performance metrics for prediction models

arxiv.org/abs/2412.10288

All I want for Christmas…is a precisely defined research question - Trials

Do carrots or mince pies improve reindeer performance? 🦌 & how can #estimands help Santa's delivery improvement research programme? Find out in ✨All I want for Christmas… is a precisely defined research question ✨
trialsjournal.biomedcentral.com/articles/10....

Yes, that's why we were thinking to never give any feedback, or to even give reverse feedback, praising errors! :D

I think it has to do with the nature of our assessments. We give 24 hour exams that students do at home. So we have to be creative and find ways to see who really understood and who didn't / is using AI. I am a bit worried about the one it did well, think I will test every time before submitting!

Asked ChatGPT to do 3 questions I prepared for various assessments in M.Sc. courses. His scores:
1) 42% (fail)
2) 45% (fail)
3) 72% (pass)
Unsure whether to be happy or worried about it.
Now the plan is to praise it for the wrong answers to train it to lead students astray if they use it.

Sample size calculations in randomised clinical trials: beyond the basics

The next delivery of our short course on sample size calculations will be on the 4th March 2025. We start from the basics and expand to more complex designs, eg factorial, cluster-randomised, MAMS. With Babak Oskooei, Ian White and Andrew Copas. Info&Registration here:
www.ucl.ac.uk/clinical-tri...

😂😂😂

[...]

[goes to Outlook and adds Chris Jarvis to the blocked senders list]

A similar one is leaving emails as unread in the mailbox. I feel dirty if I have just one there sitting unread for over 5 minutes. Meanwhile, my partner right now has 2435 unread emails, the first of which is probably the gmail welcome email.

Designs for sequences of early phase (modular) studies at University of Cambridge on FindAPhD.com PhD Project - Designs for sequences of early phase (modular) studies at University of Cambridge, listed on FindAPhD.com

I am looking for a PhD student in Medical Statistics. This project aims to develop new methods for design of early phase trials starting from monotherapy and progressing into combinations. More details are in the link. Feel free to contact me if any questions. www.findaphd.com/phds/project...

Over the 25-year history of the MRC Clinical Trials Unit at UCL, our methodology researchers have pioneered several innovative designs to make randomised cli... 25 at 25: Optimising treatments using the ROCI design

We pioneer new designs to make #ClinicalTrials more efficient & robust. Next up in our ‘25 at 25’ series: the Response Over Continuous Intervention (ROCI) design.

Find out how ROCI trials can help researchers avoid choosing arbitrary treatment options 👇 https://buff.ly/3OhgGll

Multiple imputation for longitudinal data: A tutorial Longitudinal studies are frequently used in medical research and involve collecting repeated measures on individuals over time. Observations from the same individual are invariably correlated and thus...

Hi, we just got a similar tutorial accepted for publication on Stats in Med,for the moment you can access the preprint here: arxiv.org/abs/2404.06967
It is not specifically for R but it contains R code. Also, it is for longitudinal data rather than general multilevel, but hopefully might still help!

And TBC: not taking a dig at the poor people who give away their time for free to help allocate research money the best possible way. But what am I supposed to learn from this, other than the review distribution has huge variance and hence the review process will always be hugely underpowered?

Grant rejections are tough, but can be fun as well:
Reviewer 1: You should have focused more on A!
Rev. 2: You should have focused less on A!
Rev. 3: I don't like A. Why didn't you apply for a project on B?
Decision letter: we won't give you feedback, figure out by yourself why we're rejecting.

Huge congrats to both of you!!! Looking forward to read the outputs of this research!

Improving Design and Analysis of Non-Inferiority Trials with sparse outcome data at University College London on FindAPhD.com PhD Project - Improving Design and Analysis of Non-Inferiority Trials with sparse outcome data at University College London, listed on FindAPhD.com

We are advertising a new PhD project to work alongside Michelle Clements, Ian White and myself on methods for non-inferiority trials with sparse outcome data. Funding through TMRP available for both UK and international students, please get in touch with me for more info. Deadline: Jan 13th.

Kind of a wasted opportunity to have all 9 digits exactly once in the exponent of 2.