QCovid – The UK’s COVID-19 mortality prediction model evaluated in Wales using SAIL Databank

Measures for keeping the spread of the COVID-19 virus under control resulted in many face-to-face activities moving to online and telephone consultations unless necessary.

Many individuals experienced this change in their routine interaction with clinical services such as blood tests, repeated medication, GP consultations etc. These changes may have resulted in missed opportunities in diagnosing and treatment of some groups.

It is still unknown how many people have been seriously affected by these changes and what will be the long-term impact on overall population health?

The result of the new analysis provides direct insights into how health services, such as community pharmacies, are being used across the Welsh population before and after COVID-19. This provides important insight into people’s health and uncovers the specific health conditions people are being treated for as well as supporting healthcare delivery planning and public health strategies.

A team of researchers led by Prof. Ronan Lyons analysed drug dispensing records for the population of Wales. The research team developed an interactive dashboard that illustrates the total number of medications that were dispensed to individual patients from 2016 to 2021. A simple comparison of the total number of items revealed changes in routine dispensing patterns at the start of the first COVID-19 lockdown in March 2020. In Wales, most people who were on repeated prescriptions have been given a supply for 2 or 3 months to keep required interactions under control.

This work forms part of the Con-COV initiative, Controlling COVID-19 through enhanced population surveillance and intervention, funded by HDR UK and the Medical Research Council.

In conversation with the lead authors, Fatemeh Torabi (FT) and Ashley Akbari (AA), we asked them to tell us more about their research:

Is this the first study of its kind?

FT “This is the first study using Welsh Dispensing DataSet available in SAIL Databank to generate insights on dispensing patterns over the course of COVID-19 pandemic, we show these in a near real time as part of our dispensing dashboard available at https://wdds.ml .“

AA “We also have been the first to generate a research ready data asset for Welsh Dispensing DataSet which is bringing the value and speed to further research on dispensed medication in Wales”

How did you approach the research and construct the dashboard?

FT “Our main aim was to find any signals that could tell us about a change in how people have accessed their medications through community pharmacy. For this, initially we compared 2020 (COVID-19 year) to 2019. We then looked in more details for any signal of change by comparing total numbers in each month of 2020 vs 2019. The dashboard enables users to further explore this breakdown and investigate patterns over time starting from 2016 and for 20 different types of drug chapters.“

How did SAIL enable the creation the dashboard?

AA “SAIL Databank is a Trusted Research Environment for Wales operating under the Five Safes framework. What this means for the data that we used in our dashboard is that any patient information such as name, surname, address and identifiable information was removed from the dispensing data and then replaced with a code through a process called pseudonymisation before it is available in SAIL. Also we analyse the data within the SAIL secure research platform and all our published outputs in the research article and the dashboard have been reviewed by two independent reviewer before being available outside the research environment.”

What is the potential impact of this work for healthcare delivery?

AA/FT “Many diseases get treated or controlled by drugs; therefore, we think the three main avenues of potential impact for this resource are: 1) drug dispensing patterns can provide novel insights into disease management. 2) It can also be used as a proxy measure for monitoring and evaluation of the clinical and health impact of changes in treatment patterns. 3) It is also important in identifying areas of unmet clinical need.

Full paper available here.

SAIL Databank is part of the 13 Centres of Excellence based in Population Data Science at Swansea University Medical School.