Clareece Nevill

Evidence Synthesis: From Pairwise To Network Comparisons 📈
📣Join us for an in-person course to learn and apply pairwise and network meta-analysis in healthcare evaluation through lectures and hands-on exercises!

📆 29-30 Oct
📍 University of Glasgow

🔗 Learn more & Sign up: buff.ly/5oRvObq

In 2025, many have heard of living systematic reviews - but how many are there? Are they all COVID-19 related? How are they conducted, reported, and disseminated?

My first PhD paper looks into this and more - giving a benchmark for how LSRs are actually done and shared.

doi.org/10.1016/j.jc...

Join us for a webinar with Dr Lisa Hartling, a Professor at the University of Alberta and a Canada Research Chair in Knowledge Synthesis and Translation
✏️ Overviews of Reviews: Evolving Methods, Emerging Challenges
📅 2 October
🕐 1 PM
🔗 Sign up here: forms.office.com/e/RTMzPzgBpE

A woman at a table building letters with Lego

Believe it or not - I'm not *just* playing with Lego here...

This was taken at the Young Statisticians Meeting #YSM in Liverpool last week, where I was on the organising committee. This was the intro to the session on improving presentations by focusing on stories, scripts, and slides 😎

MetaImpact: a novel tool to enable evidence-based sample sizes for future trials There is a balance when determining the sample size of a new study: with too few participants, the study may not detect an effect; with too many, some participants may undergo inferior treatments unne...

📢Work from my @nihr.bsky.social Pre-Doc Fellowship has been published in BMJ Evidence Based Medicine 🤩

It introduces MetaImpact, a user-friendly educational tool for encouraging users to consider the current evidence base when designing new studies✨

ebm.bmj.com/content/earl...

The application portal is now open at portal.unitemps.com/Search/JobDe...

Project 3: A scoping review to compare trial methodologies in neonatal and paediatric intensive care, with focus on the methodology used for outcomes. PPI meetings will shape the project direction and interpretation of results.

Project 2: Investigate innovative methods for communicating the results of a living systematic review of regional anaesthesia to patients and the public, such as interactive dashboards and/or data visualisations.

Project 1: Analyse sources of uncertainty in a probabilistic sensitivity analysis of a rare disease health technology assessment. A case study of Duchenne muscular dystrophy will be used and analysis conducted using statistical software (R/Stata).

Group Project (3 weeks): evaluate the uptake of PPI in systematic reviews and identify any common methods

UNDERGRAD INTERNSHIPS | MEDICAL STATISTICS | University of Leicester | 8 weeks starting 9th June

We have received funding from NIHR to host 3 undergrad interns! The internship will involve a group and individual project (see thread for details).

Please share around🤗
Get in contact if interested😀

Is there anything in the pipeline/already exists for R?

Looks so good and useful 🤩

This image is a data visualization titled "Based on a *True* True Story?", showing a scene-by-scene breakdown of the factual accuracy in Hollywood films. It includes nine films: Selma, The Big Short, Bridge of Spies, 12 Years a Slave, Rush, Captain Phillips, The Wolf of Wall Street, Bohemian Rhapsody, and Spotlight. Bars for each film indicate the percentage of scenes that are true (blue), true-ish (light blue), false-ish (pink), false (red), or unknown (gray). Selma is 100% true, while Spotlight is 76.2% true, with varying percentages for other films. The source is informationisbeautiful.net.

Bar chart showing the accuracy of Hollywood films based on true stories. The films are ranked by percentage of scenes true to real events: The Social Network (76.1%), The King's Speech (74.4%), Hidden Figures (74%), Philomena (70.9%), Lion (61.4%), Dallas Buyers Club (61.4%), American Sniper (56.9%), Hacksaw Ridge (52.7%), and The Imitation Game (42.3%). Color-coded for accuracy: true (blue), true-ish (light blue), false-ish (pink), false (red), and unknown (gray).

Great films fact-checked and visualised scene by scene. Is there really such a thing as a “true” true story in Hollywood? informationisbeautiful.net/visualizatio...

A case study in statistical software development for advanced evidence synthesis: the combined value of analysts and research software engineers - BMC Medical Research Methodology Background Since 2015, the Complex Reviews Synthesis Unit (CRSU) has developed a suite of web-based applications (apps) that conduct complex evidence synthesis meta-analyses through point-and-click in...

New paper📢

Myself and colleagues have written about our experiences of developing statistical software for evidence synthesis: doi.org/10.1186/s128...

Our biggest lesson learnt is that you need BOTH researchers and software engineers 😎

#EvidenceSynthesis #RSE

a cartoon mouse is laying on a blanket with the words there are no cats in america Alt: a cartoon mouse is laying on a blanket with the words there are no cats in america

The Evidence Synthesis Hackathon (eshackathon.org) is on tour this year. In early May we will be in London at UCL (more details coming very soon) & in July we are "crossing the pond" to Madison, Wisconsin, prior to the SRSM Annual Meeting (www.srsm.org/Meetings.html). Join us!

A hypothetical scenario where NMA combines all available evidence including those that have not previously been compared to each other

This was a great collaboration with Maximus, Gav and @elinatakola.bsky.social. We show how network meta-analysis, a method common in medicine, could be applied in ecology. Preprint (prior to peer-review) is here: doi.org/10.32942/X2J... #evidenceSynthesis #metaanalysis

Really exciting to see this being highlighted✨
Our suite of #CRSU apps have received such funding from CZI and its the first time that we've had funding specifically for MAINTENANCE of our apps - which is equally as important as developing new features

#RShiny #EvidenceSynthesis

Plot with areas not shaded correctly (there is a large white triangle where it shouldn't be)

Plot with areas shaded correctly

#ggplot tip to start the year! #R #dataviz

When using geom_polygon and restricting your axes, use coord_cartesian()!
For ages my polygons weren't drawing properly (left image). Turns out, using scale_x_continuous() removes data first and then attempts to draw!
All fixed now 😅 (right image)

Just finished modularising an #RShiny app for the first time 🎉

Initially it sometimes felt unnecessary - but I'm looking forward to making further improvements to the app, knowing that it'll be simpler to navigate and make changes within modules! 😎

Last year I published an interactive data visualisation for treatment ranking in network meta-analyses: doi.org/10.1016/j.jc...

A tan and white American bulldog cross muddy after playing fetch in the grass

As a first post I thought I'd keep it light and introduce the doggo!

BlueSky meet Boson - he's certainly a force-carrier (can you tell my husband studies Physics...)