Posts

Available for hire

Image
I've recently been kept pretty busy with freelance work - mostly with people I've worked with at The BMJ or in my previous life as a postdoctoral researcher. It's been so successful recently that I've decided to reduce my hours at The BMJ to focus more on freelance work. So if you have a health-related dataset or research project that you think would benefit from visual presentation, do get in touch. I'm happy to make static or interactive graphics, from one-offs to big reports / guidelines.  My contact details are in my profile. My portfolio is also available here:  https://www.behance.net/will-s-t

Visualising harms data

Image
My thoughts on where we are with visualising harms data for clinical trials (spoilers: there's still a lot to do). https://www.bmj.com/content/377/bmj.o1230 Many thanks to Rachel Phillips for inviting me to take part in her project to investigate the current state of play: https://www.bmj.com/content/377/bmj-2021-068983.full

A network diagram for our times

Image
It's been a while since I've found time to write a design story. I made some space in my schedule to talk about our recent interactive network diagram for covid-19 treatments though, as I think it's one of the most important pieces of work I've been asked to do recently. I've been wanting to have a go at improving the display of network meta-analysis diagrams for a while now... The article is available on bmj.com: https://blogs.bmj.com/bmj/2020/08/26/will-stahl-timmins-a-network-diagram-for-our-times/

Video: "how we make a graphic"

My infographics for The BMJ often take a while to develop. There are a lot of decisions to be made about what information to present. Even with a full-page graphic, there's a lot of backwards and forwards between myself, the editorial team and authors, about how best to represent information from a published article in visual form. In this case, we produced 52 different versions of a graphic on diabetic foot, which I've animated together here. Find the final version here:  http://www.bmj.com/content/359/bmj.j5064

Almost impossible cancer spaghetti

Image
NICE guidelines, produced by the National Institute for Health and Care Excellence, are detailed, and sometimes lengthy documents. They are an attempt to summarise all available evidence on a health topic, supplemented by expert opinion where evidence is not available. The BMJ has recently been publishing quite a few summaries of these NICE guidelines, aiming to provide doctors with more accessible, easy to use resources. I've been making a few infographics too - mostly static ones that people can tear out and stick on their consulting room wall, or print from the website. One of the first ones I worked on was a bit special. The NICE guidelines on referral of suspected cancer were updated earlier this year, and I was asked to take a look at the tabulated data and try and turn it into something more visually appealing. I was presented with this (on a Wednesday evening)  - a series of tables detailing the advice on what to do when children present with some key symptoms:

A web of influence

Image
A collaborative project from earlier this year that I enjoyed working on was The BMJ's investigation on the sugar industry, and links to UK government advisory panels. For this project I was working with a journalist, Jonathan Gornall, to co-create the graphic. The project started, as so many projects do, with several long tables of data - in this case Freedom of Information act requests, and publicly declared conflict of interest statements from different organisations: I initially began to explore a force-directed node diagram to try and see an overview of these complex links between organisations and individuals. I first tried a d3 layout, by d3noob:  http://bl.ocks.org/d3noob/5141278 https://dl.dropboxusercontent.com/u/56571574/sugar/index.html This version started to reveal important hubs, like the MRC and Susan Jebb, but it wasn't easy to see the detail beyond this. I edited the code, to freeze the dots in place when they were dropped, allowing me more c

prostate symptom checker

Image
I've recently started a new job in London at the BMJ (formerly known as the British Medical Journal). It's a new role they've created for an interactive data graphics designer. I'll be making graphics, mostly in Processing to begin with, which will be collected in a free archive on the BMJ site. Since my first interactive has just gone online, I thought I'd put up a quick development post about it. Those of you familiar with the development posts on this blog will recognise the workflow I've been using - starting with pencil sketches and using these to discuss ideas with subject experts, then working up the graphic in Illustrator. However, as I'm doing more interactive graphics from now on, I've been learning to use Processing . This means that I've had to make changes to the workflow, adding in a quick and dirty functionality test so that I can iron out issues before committing myself to the design. I started this project as I do with many o