Steff Lewis

Professor of Medical Statistics, University of Edinburgh

Can you tell us about your journey to your current role?
I was always good at maths, for as long as I can remember. My Mum said I used to add up the numbers on playing cards for fun (laughs).
I did a maths degree and at the end, I met with the careers service and they were rubbish! They said ‘do you want to be an accountant?’ and I said ‘no’. I went home and said to my Dad that I didn’t know what I should do – I liked maths, biological sciences, stats; and he said ‘why don’t you do what I do?’ So I did – he was a medical statistician. That sent me off down that route.
My first job was with a Clinical Research Organisation based in Reading. I was there for two years and then did a Masters’ in Medical Statistics at Leicester, because I realised I needed to know more to develop and take the next step.
There was an opportunity to work in Newcastle in the field of pharmacovigilance. My boss there was Professor Sir Michael Rawlins – he was the Head of Committee on Safety of Medicines. He was a very interesting man to work for – he used to roar with laughter and slap me on the back if I said anything funny. He encouraged me to do a PhD.
My fiancé and I wanted to be in the same city – we were in a distance relationship. He applied for various things and didn’t get anything in Newcastle. Then we both got jobs in Edinburgh, so that is where we ended up. My first job in Edinburgh was with Sheila Bird (Gore at the time) and she was researching HIV risk and illegal drugs among prison populations. The research involved interviewing prisoners about their drug and sexual habits, which was fascinating for a 20-something-year-old statistician! (laughs) The funding for that then came to an end and I went on to work with Professor Joanna Wardlaw in clinical neurosciences.
I’ve always been really interested in clinical trials and meta-analyses of them. I naturally drifted towards the big clinical trials that were happening in the clinical neurosciences department. There was also a discussion in Edinburgh at the time about having some sort of generic clinical trials unit, and I was at those meetings. Now Edinburgh Clinical Trials Unit employs 60-70 people, and I’ve been here since it was created!

What is a typical day at work for you now?
My typical day involves a wide range of studies, and a variety of statistical issues and management issues. No two days are the same. The are three statistical professors who split the senior statistical responsibilities, and a team of 11 statisticians.
We are working on a range of projects – about 50 at any one time. A third of them are regulated drug trials, so in theory, the Medicines and Healthcare products Regulatory Agency (MHRA) could walk in the door any time and audit us. For those studies, there are legal requirements and they are strictly run, so this keeps us on our toes. We work across a huge variety of medical areas, which keeps work interesting.
In addition to the day-to-day work and management, I sit on steering and data monitoring committees for other people’s trials. On these committees, you’re checking trials to see if there’s anything dreadful or anything spectacular happening so you can intervene appropriately – however, it’s mostly boring, in a good way!
We work on methodological questions as well. I’ve done a lot of work on data sharing, in particular as how we can share data at the end of a study – I have a PhD student working on anonymisation of data. If you take the names and addresses out of a dataset, you can perhaps still identify a person by gender plus date of birth. But the more data you take out, the less usable the dataset is.
I’m also co-director of Research of the Usher Institute. It’s a job share, which I love because I’m a collaborative person and like to do things with colleagues, I enjoy bouncing ideas off other people.
And I’m on a UK funding panel, get involved in teaching, and numerous other things. So it is definitely never boring.

Do you find your work supportive of women?
When I first started, there were a few sexist people about. But that was a long time ago, and things have changed for the better. I’ve always thought that you should carry on being who you are, treat everyone the same, and get on with life regardless. So yes, I find my workplace supportive of women.
I don’t have children myself but people say I am very nurturing towards my team. I like my workplace to be a good place to work. I’ve heard people say that they think that our workplace is accepting and supportive in general, and understanding of life and the things that it throws at you – whether that’s maternity leave or sickness or anything else.
I’m a huge believer in work-life balance. I’ve worked four days a week for over ten years. It changes your life, it gives you headspace – I come back in on Monday morning after a proper rest. I don’t work on my days off and I’m very strict with myself about that. I do still think about things and mull over matters, my brain is ticking, but I enjoy my days off. I think my work gets just as much out of me. Work piles up, of course, but that would happen if you were full-time. Stuff outside work is important.

What do you do with your time outside of work?
I make jewellery. I go to a silversmith’s workshop, and one evening a week, I get to hit things with a hammer. I also started playing the guitar last year.

What has been the best opportunity in your career thus far?
Blimey, that’s hard.
I think being in the position to grow the trials unit here. Looking back now at how far we’ve come, it’s incredible. I’m not the sort of person to say ‘I’ve done this, I’ve done that’ but it’s a huge sense of achievement that we’ve created the trials unit together.

We give you permission to blow your own trumpet!
A couple of years ago I was involved in a trial with Steve Cunningham, a respiratory physician. The trial was of whether children with bronchiolitis could be sent home when their oxygen levels were a little bit lower than was currently done. We showed that they could safely be sent home a day earlier.

Yes, you can really see the impact of that.
The thing that I really like about what we do is working without having to make a profit. The pharmaceutical industry is working hard, doing brilliant stuff and changing people’s lives, but where profit can be made. But for the questions that need answering, but where profit can’t be made, there’s no money for a company to make. Those are the questions that I like to answer. That is what gets me out of bed in the morning; the idea that medical statistics are improving care for people and making a real difference to people’s lives, in a world where ethics and good practice are the reason for doing what you do, not profit.

Can you tell us more about the data sets you work with?
The majority of our data is directly recorded from patients in to our data sets. We also have some studies ongoing where the data is from routine NHS databases.
I spend a lot of time designing new studies. I like to design studies in ways that normal people can understand, rather than using whizzy black-box techniques where it is unclear to others what has been done. Increasingly in data science, there’s a ‘black box’ where the data gathering and analysis takes place and it’s harder to articulate to people what the actual processes are. I’m not rubbishing it, there’s a lot to be said for these complex computational processes, but I’m drawn to the clear and simple findings from clinical trials, where it’s obvious what the data are telling you.
My immediate reaction to an analysis of study results is to ask ‘do you have a table or graph that shows the data?’ I like results to be displayed in a way that anyone can see the effects, and can really understand what has been done.

A really refreshing approach! So what are you looking forward to?
I look forward to continuing to support a structure and culture of nice people to work with. And an endless variety of new interesting questions to answer.

Is there anything you’d recommend to women and girls who’d like to be you?
Don’t be afraid of talking to people. We have an incredibly accessible internet and you can easily find out what people do and contact them. If I was 15 or 16, I wouldn’t want to talk to the version of me now because I’d be scared of me, but I shouldn’t be.

And don’t underestimate yourself. I never thought I’d get to where I am.

Do you have a fun fact about yourself?
There’s my creative side, my jewellery making – I enjoy playing with 800 degree ovens. Oh, and as a student, I used to organise p**s-ups in breweries. I was the president of the real ale society during my undergraduate years! (laughs)

Excellent fun fact, one of the best we’ve had! Do you have a hero or heroine?
I don’t have a single hero or heroine. There are huge numbers of people that have influenced me and helped me get to where I am. My heroes and heroines are the people who don’t have their names up in lights, and never blow their own trumpets, but are there when you need them.

Image of Professor Steff Peebles
Picture by Lesley Martin, interview by Poppy Gerrard-Abbott

That is what gets me out of bed in the morning; the idea that medical statistics is improving care for people and making a real difference to people’s lives

Read the latest Women in data

Image of Natalie Duffield

Natalie Duffield

CEO of InTechnology Smart Cities Ltd

Photo of Hoppers Society Members

University of Edinburgh Hoppers Society

Image of Samantha Rhynas

Samantha Rhynas

Head of Data at Effini & Girl Geek Scotland Leadership Team & PyData Edinburgh Organiser