Dr Helen Purtill talks about her interdisciplinary research as a statistician and member of the Health Research Institute for International Women’s Day 2024.
I am a statistician working with data in interdisciplinary research, where diverse perspectives, expertise and methods can address important societal issues.
There are always opportunities and challenges when working with data, from biases to missingness, sample sizing to reliability and validity. Dizzying advances in technology and statistical methodologies have changed the landscape, but it remains as important as ever to carefully design studies, identify feasible and relevant research questions and to collect appropriate and useful data.
Data visualisation is so important in today’s fast-moving society, but graphs were crucial to decision making long before the age of computers. I am always inspired by the wisdom of Florence Nightengale’s famous 1858 graph from “Notes on Matters Affecting the Health, Efficiency, and Hospital Administration of the British Army” - a powerful picture depicting the extreme mortality rates due to unsanitary hospital conditions during the Crimean War.
Mathematician and statistician John Tukey famously said, “the best thing about being a statistician is you get to play in everybody else’s backyard”. This, I completely agree with, especially within the Health Research Institute (HRI) where collaborative research thrives. In recent years, I’ve had opportunities to work and publish in a wide range of research areas including chronic pain, sleep patterns of athletes, cognition in older adults, injury surveillance in rugby players, biomarkers, medical devices, randomised control trials, birthweights of babies, systematic reviews, meta-analyses, training load in sports, medication, and healthcare utilisation.
In my experience, each dataset is different with its own characteristics and story to tell. An area of statistics that I work extensively in is latent class analysis, a person-centred methodology that seeks to identify previously unseen patterns and subgroups within populations. For example, by modelling longitudinal data from The Irish Longitudinal Study of Ageing (TILDA) it was possible to identify biopsychosocial risk classes linked to the development of pain in older adults. Findings from this study highlight the importance of considering an older person’s systemic health, social circumstances, and overall life in the treatment of pain. A recent publication examined biases, differences in reporting styles and social disparities in pain in older adults by modelling five waves of TILDA data. The results provide important considerations for estimating pain in populations using longitudinal data. In these studies, I worked with talented PhD students, fellow statisticians, experts in chronic pain and international collaborators.
I’ve been privileged as a member of the HRI and a founding member of the Ageing Research Centre (ARC) to work with fantastic people, dedicated to research studies that help people live healthier lives. As a member of the Mathematics & Statistics department I work with excellent researchers and PhD students in UL, nationally and internationally. I really enjoy connecting people and data to deliver insights into the world we live in.
As a woman and a mother, balancing all the aspects of work and family life can be tricky. I have been lucky to have had strong female role models throughout my life, as a daughter of a former Irish hockey goalkeeper and the granddaughter of a horse trainer I was brought up to believe that women are strong and can work in any discipline. In my academic life, I’ve been lucky to have great role models in more senior female academics, and I always try to support younger colleagues and PhD students on their academic journey.
Advice for women in research – together we’re stronger! It’s important to find what you really enjoy working on and to find good people to work with.