A statistician in a clinicians world
By Aoife O’Neill
As a statistician, we often find ourselves surrounded by academics outside our discipline, but is this a good or a bad thing?
I am a third year PhD student and this year starting working with the Professor of Biostatistics of the Medical School to offer statistical support to the clinicians carrying out research. I truly believe that working with clinicians has given me an insight into the world of research that I would not have gained through my PhD alone. Don’t get me wrong, it hasn’t always been smooth sailing and for the remainder of this post I will discuss some of the pros and cons of working with clinicians and undertaking side projects throughout the PhD.
As a PhD student we may find ourselves consumed by the ‘one project’ that is to be the focus of our research for the first four (or so) years of our early academic lives. However, I find it increasingly helpful to have a side project to offer some time away from my PhD and to allow time to explore other methodologies and applications. It has been my time spent working with the medical school where I have gained the most access to such projects.
An example of some of the projects I have worked on recently are
- a proportional meta-analysis of substance abuse and psychological problems of the Irish prison population;
- a survival analysis of a specific replacement knee brand for a sample of patients treated by a top Irish arthroplasty surgeon;
- a morphometry data analysis of expectant mothers over the course of the three semesters of pregnancy;
- an analysis of a psychological intervention programme for patients with chronic pain; among many more.
An added bonus of working on such a large number of other projects is the publication count beings to rise early in the academic career.
One point of working with clinicians that is either an advantage or disadvantage, depending on how you look at it, is the fact that you are often the lone statistician working on the project and thus hold a large responsibility. The entire process of data cleaning, data analysis, and report writing can be difficult if you cannot discuss your ideas with someone with a sufficient level of statistical expertise. I am fortunate to have the guidance of an excellent statistician; however, I can see how this role could be isolating for some.
At times through this experience, difficulties have arisen when the clinician’s level of statistical knowledge is considerably low and their main concern is whether or not they can see the magical p<0.001 – which is not always the end result. If an experiment has not been designed rigorously, it is still the job of the statistician to tidy up the mess of data. It can be challenging to explain the insights from the data and the methodology employed to empower the clinician to make informed decisions on the protocol under investigation. Communicating complicated statistical or mathematical concepts of a project to a general or non-expert audience is a vital transferable skill. Clear communication highlights the benefit of employing statistics or mathematics to an applied project and strengthens the overall collaboration.
This experience, while challenging, is very rewarding and something I would recommend to all PhD students.
Aoife O’Neill is a PhD student in the Department of Mathematics and Statistics at the University of Limerick working with Dr Helen Purtill and Professor Cathal Walsh.