Safia Nasuuna blogs about the project she’s been working on as part of her summer placement with BREATHE, through the 10000 Black Interns programme. Safia’s internship was based within the Usher Institute at the University of Edinburgh.
While working on this internship, I have been analysing COVID-19 related data to look at the impact of pharmaceutical and non-pharmaceutical interventions on COVID-19 infection, severe illness, and deaths. The data I have been using has been taken from Our World in Data and is publicly available. The (UK) data dates to the first case of COVID-19 in the country and tracks different variables throughout the pandemic, such as new deaths, new cases, and hospitalisations.
To analyse this data, I used coding techniques in the programme R, which is something I previously only had a basic knowledge of, so it took me a lot of trial and error to figure out how to use. In this process I have become a lot more confident using R and coding. I used the coding to clean up the data and retain only the information I needed (such as filtering the location to the UK), and create visualisations of the data.
One of the main things I focused on to explore the impact of the interventions was creating a time series graph: this is a time/(variable) plot that shows how the variable changed over a period of time. The type of time series I used was an interrupted one, where a single intervention was placed in the middle of an 8 week period. The benefit of this is that it allowed me to see how that single intervention impacted the modelling of the variable.
This kind of research would have a huge impact on evaluating how the pandemic was handled, and how best to go forward. With this type of research, we can find out whether previous interventions had a positive or negative effect on controlling the virus then decide which ones to continue putting in place.
The data contained information from all around the world, so it would also be beneficial to look at countries that handled the pandemic more effectively and analyse the effects of the interventions they put in place. We could then adapt their interventions and apply similar techniques to the UK to increase efficiency. The impact of this research would be key to getting the virus under control. In the long run, it could reduce new cases, new deaths and severe illness caused by the virus.
It's been really interesting to work on something so relevant, and to see the real life implications of health data research.
Watch: Safia describes her highlights and key takeaways
https://media.ed.ac.uk/media/1_f5w85hff
HDR UK Black Internship Programme
More about the Black Internship Programme from Health Data Research UK
